Transcript

First developer job at age 38 with lawyer turned software engineer Zubin Pratap [Podcast #203]

From freeCodeCamp.org · Episode page · Watch on YouTube

· 1 h 13 min · 15,602 words · 163 timestamped sections

This is the full, timestamped transcript of "First developer job at age 38 with lawyer turned software engineer Zubin Pratap [Podcast #203]", an episode of freeCodeCamp.org published January 9, 2026 (1 h 13 min, about 15,602 words). It mentions 10 people, works and ideas, including Free Code Camp, Google LLC, Quincy Larson, Zuben Proup, Johnson Space Center.

What is discussed, and when

  1. – 10:00: Free Code Camp, Google LLC, Algo Monster, Artemis mission
  2. – 20:00: Free Code Camp, Google LLC
  3. – 30:00: Free Code Camp, Google LLC
  4. – 40:00: Google LLC, Free Code Camp
  5. – 50:00: Free Code Camp
  6. – 1:10:00: Free Code Camp
  7. – 1:13:12: Free Code Camp, GitHub, Zumin Pr

Notable quotes from "First developer job at age 38 with lawyer turned software engineer Zubin Pratap [Podcast #203]"

  • Welcome back to the Free Code Camp podcast. I'm Quincy Larson, teacher and founder of Free Code Camp, and I hope your 2026 is off to a great start so far. I'm hyped to be interviewing Zuben.
    about Free Code Camp · at
  • Here in the US or able to travel to the US. absolutely should do this. Over the holiday break, I took my kids to Johnson Space Center in Houston and it was just insane. I loved…
    about Johnson Space Center · at
  • Finish listening to this podcast. This podcast is made possible by a grant from Algo Monster. Algo Monster is a platform that teaches data structure and algorithm patterns in a…
    about Algo Monster · at
  • Some drills, regularly. Yeah, absolutely. Well, don't neglect those fundamentals, everybody. Zumin Pr it's been an absolute pleasure having you back on the podcast after seven…
    about Zumin Pr · at
  • Francisco this is when I was learning to code by the way I hadn't really gotten any good but you had that at the GitHub headquarters in San Francisco yeah the J that's it…
    about GitHub · at

Who and what is mentioned in this episode?

People

  • Quincy Larson — Person · 1 mention · first at
    Welcome back to the Free Code Camp podcast. I'm Quincy Larson, teacher and founder of Free Code Camp, and I hope your 2026 is off to a great start so far. I'm hyped to be interviewing Zuben Proup, a corporate lawyer who taught.
  • Zuben Proup — Person · 1 mention · first at
    Founder of Free Code Camp, and I hope your 2026 is off to a great start so far. I'm hyped to be interviewing Zuben Proup, a corporate lawyer who taught himself programming in…
  • Zumin Pr — Person · 1 mention · first at

Companies & orgs

  • Free Code Camp — Company · 10 mentions · first at
  • Google LLC — Company · 8 mentions · first at
    Himself programming in his late 30s using free code camp, then landed a software engineering role at Google.
  • Guns and Roses — Company · 1 mention · first at
    Popular. And then for this week's song of the week during the road trip, I listened to the entire Guns and Roses catalog, which I hadn't listened to all the way through since I…
  • Algo Monster — Company · 1 mention · first at
  • GitHub — Company · 1 mention · first at

Concepts & topics

  • Artemis mission — Concept · 1 mention · first at
    Experience created by Tom Hanks and a couple other people about this year's Artemis mission to the moon. this is an incredible museum. We spent seven hours there. It wasn't…

Places & events

Full transcript with timestamps

Welcome back to the Free Code Camp podcast. I'm Quincy Larson, teacher and founder of Free Code Camp, and I hope your 2026 is off to a great start so far. I'm hyped to be interviewing Zuben Proup, a corporate lawyer who taught himself programming in his late 30s using free code camp, then landed a software engineering role at Google. First, let's jump to some community news. 2025 was a pretty awesome year for the Free Code Camp community. We shipped 129 free video courses right here on the Free Coamp YouTube channel if you're watching on YouTube. We also shipped 45 fulllength books and handbooks. These were all free on the Free Coamp publication freeamp.org/news.

And we shipped 452 programming tutorials and on math, programming, computer science and 50 episodes of this very free code camp podcast where I talk with devs. So really productive year and at the end of the year we launched version 10 of free code camps JavaScript certification, Python certification, relational database certification and responsive web design certification. And you can earn all these certifications for free and put them on your LinkedIn, your resume, your CV, your personal website, which I hope everybody out there is getting a personal website. As a developer, it's important to have your own square of the internet for yourself and not be completely beholdened to platforms. And uh free co also shipped the Spanish curriculum and we shipped a Chinese curriculum, too. This is just the beginning of the work we're doing on teaching foreign languages. We have big aspirations and it's going to take a

couple years to get all of this live that we're developing. But if you've ever wanted to learn Spanish, you can learn it interactively. You'll learn pronunciation introductions numbers and more. There are more than 200 steps live. And it's got this cool animated interface where you're talking with different characters. And so it feels like you're going to work at as a software engineer in a Spanish-speaking country. It's going to be a lot of fun. And we've already got a little bit of A1 live and we're going to get A2 and B1 live hopefully some point in 2026 2027.

So and if you want to study the Chinese uh uh so this is live on free coamp.org I have a travel tip for everybody who is here in the US or able to travel to the US. absolutely should do this. Over the holiday break, I took my kids to Johnson Space Center in Houston and it was just insane. I loved it. We visited the mission control center that NASA used during the Apollo missions to the moon and they had this awesome multimedia experience created by Tom Hanks and a couple other people uh about this year's Artemis mission to the moon. Uh this is an incredible museum. We spent seven hours there. It wasn't particularly expensive. Just make sure you book your tickets to Mission Control a few months out uh just because it is extremely popular. And then for this week's song of the week during the road trip, I listened to the entire Guns and Roses catalog, which I hadn't listened to all the way through since I was a kid. And

that brings me to this week's song of the week, the 1991 epic Estranged. This is perhaps the most mature sounding song that Guns and Roses has ever done. And it's got really introspective lyrics and a series of very tasteful, very emotional guitar solos and some great drum fills, too. Check it out after you finish listening to this podcast. This podcast is made possible by a grant from Algo Monster. Algo Monster is a platform that teaches data structure and algorithm patterns in a structured sequence. You can learn uh a whole lot of approaches to technical interview questions more systematically. Their curriculum covers patterns like sliding window, twoo pointers, graph search, and dynamic programming, helping you learn each pattern once, then apply it to solve many problems. Start a structured interview prep routine at algo.mmon/freecode.

Support also comes from the 10,338 kind folks who donate to our charity each month. Join them and support our mission at donate.freeccoamp.org. And if you want to own this awesome free co shirt, not this one that I'm wearing that's all sweaty and stinky, but a fresh one printed by hand made here in the US by Raphael Hernandez who runs the Spanish free co podcast. He has a screen printing company and uh he will print it and ship it to you 20 bucks anywhere in the US. If you're outside the US, just grab our assets from our asset repo and screen print it yourself or take it to a local t-shirt shop. You can do that for your own personal non-commercial use.

But 20 bucks shipped anywhere in the US. Check it. Then uh I am hyped to introduce Zuven Prown for many years, hung out with in person both in Australia and here in the States in San Francisco. He's a software engineer and manager from Melbourne, Australia. And after two decades working as a corporate lawyer, he taught himself programming using free codecamp.org. Within two years, he had landed a job as a software engineer at Google. Now, it wasn't his first developer job. He got a developer job and then he got Google as his second job, but it's still incredibly impressive for somebody in their late 30s to be doing this. Uh, and we talk about how the tools are making programming easier, but other parts of being a developer are getting harder and how 2009 to 2022 was not a normal job market for devs and how devs are adapting to the changes in the job market. And we also talk about this quote from him. The purpose of

communication is to be understood and other lessons Zuben's learned over the years. Zuben for top, welcome to the free coke camp podcast. Hi, Quincy. It's great to be back. It's great to see you.

Yeah. You were here in like 2018, I think. Yeah. 18. Yeah. It was like six years ago. Seven years ago. So, we're thrilled to have you back on the podcast.

Yeah. Yeah. That's right. That's right. Yeah. Yeah. Yeah. Yeah. And we're going to get into your developer journey to recap that for anybody who doesn't remember from seven years ago or missed that podcast. Uh but I want to start be you are uniquely qualified to answer this question. So of course LLMs and you know other AI technology, robotics, uh computer vision, stuff like that, they're changing a lot but you have had the opportunity to work as a corporate lawyer. You've done management in industry and you've worked as a software engineer and you have a lot of perspective in to how these tools are changing all of these fields and changing like the nature of work itself.

uh how are they changing the fields you've worked in? Oh well um I mean you you can see it everywhere. So ultimately anything that's done on a computer um is data that um LLM's machine learning algorithms can access right and then there are patterns that emerge out of the way we use computers whether it's spreadsheets all the way to X-rays right now all of these things the moment we adopted digital technology we opened the gateway for uh machine learning algorithms to apply to the data that goes through the digital technology. So law for example, I mean there's no no secret lot there's plenty of legal startups out there helping with you know large large volumes of text. Now law is a very linguistic languagebased profession very well suited to LLMs. Um you know even things like management consulting, accounting um etc use large language models now more and more because it's a very efficient way to

process information at a speed and scale and efficiency that we can't do as human beings. Like there are cognitive limitations we have. So that's fantastic because frankly it was a lot of the rubbish work. Heck, even software engineering, I do more projects now than I did even at the start when I was learning. Right now part of that is because I'm faster. I know more. I'm able to make better judgment calls. Um but also part of it is a lot of the crud lower level setup stuff, all the really painful stuff I don't need to do. I can supervis the machine doing that, right?

Um so there's that in in medicine. I see it all a lot as well because there's lots of data that's processed as part of medicine. I'm not just talking about medical research. I'm talking about day-to-day patient care. It's all going through a computer system somewhere. The moment it touches a computer system, it lends itself to uh to AI, machine learning, whatever you want to call it, you know, large language models or even, you know, non- language model AI, right? Like that's what my radiographer recently told me. You know, they use AI now to to scan far more X-rays than they can as a human being and they exercise the judgment layer over that as well. So yeah, it's good. It's a fantastic technology. Uh scary and difficult initially for sure, but you know, I think I'm very optimistic. Yeah.

And you recently broke your finger uh like a few weeks ago. How's it doing? By the way,

it's it's better. It's still attached and I I've got my mobility 80% back. Thank you for asking.

Awesome. Yeah. I hope you get back to 100% mobility and I hope it's not too painful of recovery. But that experience you you told me that that introduced you to how it's not just developers that are scrambling to make sense of how to use these tools and how to not use these tools, but it's also physicians and people throughout like the healthcare field. Um can you talk about that a little bit like what you observed?

Sure. And and I've been actually observing this for almost half a decade. So when I was still at Google um you know transformer technology had come out and people were already starting to talk about how this is going to affect knowledge work, right? And so more recently when I broke my finger in uh in November um yeah pretty much all the healthcare practitioners I spoke to when they found out I was in software engineering and I was a software engineer they wanted to talk about AI. Um and my surgeon for example uh believes that he would rather that you know my surgeon did say that he believes that you know robotics will take over a lot of the engineering work because you can scale it better it can be much more accurate. Um, of course it requires all sort of changes around how processes are done. But he believed the role of the surgeon would then be more of the pilot and supervisor, the operator, um, rather

than the person actually doing the cutting, the incision, you know, going in, etc., etc. But he would supervise and that would also give him the ability to do more remote healthcare and and you know things like that. So he was very optimistic. My optometrist unrelated to the finger but you know part of the same medical suite of things that got done uh was also very interested because obviously when he's looking at retinas and you know eyes and examining them a lot of it is just data and it's just numbers you know uh testing your vision what that means for the health of the eye all of that he believes that AI tools are critical for him to learn my radiographer who did the X-rays um she was well she said quite frankly the AI can do a lot more than anyone any one radiographer can and much faster you know what takes a human being about five years of experience um the AI can sort of ingest and do in in in a in a week.

Um but so that's not to say it's going to displace them necessarily. It's just going to say that all of these professions feel that their value is going to go up the chain. You know, there was a time when when you had your horse and buggy, you kind of know need to know how to feed the horse. You know, you need to know know how to change a cartwheel or whatever it is. And now we all drive cars that we have no idea how to repair.

Right. Right. A lot of us don't have any. I don't really have any. I know how to change my tire and that's about it. Change my oil.

That's about it. I don't even own a car anymore, right? Because I just Uber everywhere now. So, you know, what you end up doing with the technology changes. Um, but the fundamental utility of the technology doesn't change, which is, you know, in that that example transport or but it's medicine or software engineering. It's the creation of software that's robust and scalable, usable and likable. Um, and the role of the software engineer in that process will change.

Yeah. and we're going to get into how you think it's going to change. But first, for those who haven't heard your story, it is a banger, alltime banger story. So, I think you're like 36, 37 years old, and you decide as this corporate lawyer nearly two decades in the law helping deal with like giant contracts and stuff like that. I don't know what corporate lawyers do. Uh, you decide you want to learn programming. Uh, and then you of course went on to get your first developer job about a year later. And then after that you got a job as a software engineer at Google and you've worked as an engineer and as an engineering manager since. But I want to go back to that time when you decided to learn programming because that takes some audacity to be mid-career and to basically say I'm going to start over in a completely different field. Like can you walk us through that period of your life?

Yeah. uh you know and I'm going to compress into like 30 seconds what was effectively almost 5 years because you know I'm in almost in my mid-40s now so we're talking almost a decade ago where the story started I I really started try my first attempt at learning to code I remember distinctly was 2013 and then I promptly gave up and then again I tried in 2014 and then 2015 and then 2018 and on the fourth attempt I succeeded but for you know it was a long and painful journey and just to to remind us all of what the world was like back let's say 2013 uh you know Uber was starting starting to get mainstream. Airbnb was starting to get mainstream. We were still very early in the mobile phone revolution. WhatsApp was starting to be, you know, be quite widely used around the world, but there was no real Instagram penetration at that point of time. You know, YouTube wasn't what it is now. Like the world was a different

place. Spotify wasn't what it is like none of that existed, you know, and so very different world. And uh come 2018 when I actually met succeeded in learning to code. And my definition of success is um I' I'd learned to code something, you know, over all those years. but uh not good enough to get a job. And to me that was the the the the test of have you learned to code as well? Will someone hire you? Right? Because you can actually know how to code without anyone hiring you. But you have to decide what your goal is. There has to be some objective metric for success for yourself. And that was my metric. Um but I only really seriously started focusing on the job market in in 2017 2018 um because I had really bad goals before that. So I sort of achieved very

bad goals. What do you mean your bad your goals were bad? My goal was I want to learn to code. And I I honestly believe that that's a very vague goal.

Yeah. Too vague. You know, I think you talk about this in your book actually, right? About uh about having a very specific set of intentional goals and about actually the it it's going to take you longer than you realize u to get all those skills. But you really have to sort of repeat and repeat and repeat and go deep. And it's frustrating. And I think people just think we can learn to code because we see some, you know, 30 second short or something about, oh, I learned to code in three months. But what does that actually mean? You know, writing syntax is not definitive.

Passing some random DSA test is not definitive, you know, on that. Like your goal has to be specific for me. For a long time, it was I want to learn to code. And I started Quincy crazy. I started trying to my starting point was Android programming. Like

that's how little I knew where I recommend people start. No, you know, it was two attempts, two years, and I couldn't figure out anything because there's so much foundational knowledge required, you know. Um, and then I, you know, I, yeah, I switched to other things. I tried Jav, I tried data science for a while, couldn't figure it out. And it was only when I had to start right at rock bottom, you know, HTML, CSS, then JavaScript. And this is where, you know, free code camp's curriculum because what it helped me see was, hey, there's a structure. the the information has always been there, but without a structure and a sequence, it's like trying to teach a kid calculus before they know arithmetic, it's not going to work now. You could say, I want to learn math, right? Okay, that's a pretty big subject and you got to do it in the right sequence, you know?

Right. Yeah. And that's like if you look at software and even even the hardware layer, it's it's just a stack of layers of abstraction. And that's why a lot of computer science curricula, they just start you at C. like okay we're going to learn C all right like that's what almost everything is built in and then you'll learn Python later or learn JavaScript later but like pre camp doesn't start that way because frankly it's

demotivating and it's very difficult to program so we we start higher up on the stack and then we go down but then we go back up you know so

and I know no C so clearly you don't need it to to to have a job you know I know zero

you don't need it it's a nice to have it depends on like how high performance and you know stuff you need to But uh it doesn't hurt to learn it. And Fre has lots of C courses if you want to learn. Uh David Men, whom I recently had on the podcast, Harvard CS50 prop, he still recommends people learn C and he does teach it in CS50.

Um yeah, but but but yeah, I don't know C. Like I'm outing myself as not a true Scotsman, right? I'm not I don't know C really well. But no, I I have very limited experience with it.

Yeah. And you know what one thing I will say is that I think a lot of people tend to have this is the mental model which I flipped that really helped me and you know has helped a lot of my mentees over the years is I get people to identify what they think they need for their goal and then I will systematically try and strip it back because what we tend to do is we tend to want to acquire a bunch of knowledge without analyzing its utility as judged against our goal. There's no end to that Quincy. There is no end to knowledge, right? There's no end to the amount of information. But time is so finite that per unit of time, if you're not directionally really valid, that knowledge has very little utility to you, you know. Um, and I think people really need to get clear. And I I call it the minimum effective dose when I'm mentoring folks is, you know, what is the minimum effective dose to get you to

your result? It's it's from pharmarmacology, right? Like you can throw a bunch of drugs at a at a problem, but you will probably hurt yourself more than you help yourself if you do that. You're probably better off looking for the minimum effective dose to fix that problem. Um, and it's the same thing with with acquiring complex deep skills.

Yeah. And this brings me to a topic that I'd love to talk with you about. So you have observed in our conversation uh before we started recording that uh you know libraries are in a lot of countries like in a lot of countries you can got get on a bus or you can maybe even walk to a library where there will be tons of books on basically any subject including mathematics. Uh you can get lots of programming books. You can get books on uh any skill that you want to acquire. Right? And yet very few people just go to the library and teach themselves physics for example like right why is that

because I I was listening to a podcast you know farm street I think it was uh recently where you know I think James Clear was talking about the human brain seems to be very wired to look for shortcuts. Um, we want the magic bullet, the one secret, the one tip, you know, we we we really want the the outcome without the process. Um, and so I think a lot of people don't do the hard yard. I mean, let me ask you this. Do you remember how you talked about when you were learning, you locked yourself in a closet, I think, right? You said

for weeks. Yeah. I had to force myself to do that cuz like you, I started and stopped repeatedly and I would get like shiny object syndrome. I'd go on hacker news or wherever and I'd start reading about like all this doom technology and I'd be like maybe I should learn that

but I had to learn the fundamentals right and and I'm willing to wager that this was in a world where there was less noise and less information than there is today because yes there was hacker news yes there was Reddit um you know but these are still very early on we didn't have the the delusion of video based social like it was still a noisy world but somewhat less noisy and even then you had to lock yourself do the really hard thing which is eliminate things to do and just literally close your world down to such a narrow level that you had that it implicitly meant focus by excluding everything else.

And I had to own a smartphone at the time. Right. Right. I had a like a a Nokia dumb phone thing.

So I I would when I went to the bathroom, I was thinking about stuff. I wasn't looking at my phone,

you know. So even that is like an additional challenge now because when I go to the bathroom, I'm like, "All right, let's let's see what uh going on on something." So something so, right? Yeah. There was nothing to scroll back then.

I agree. And so for you, the big thing then would have been like it almost doesn't matter what you ended up studying because look, it's the same whether you do C from this textbook or that textbook. The C is not different. It's the same thing, right? What matters there is the sheer grit and discipline to stick with one thing long enough till you then know what the next thing is you should do and at the same time eliminate all other distractions and shiny objects. Now that has not very much to do with the library or the you know whether it's free code camp or something else that has to do with just the sheer amount of focus and will and drive you're willing to have and then of course you want to choose the right tool the right textbook the right you know whe it's free code camp you know um or whatever it is people even if it's not coding uh I mean I'm selftaught of the guitar as well you know the one rule I

had for 10 years was don't buy a second guitar right because it's so I have exactly one guitar and exactly one bass and I've spent hundreds of hours on both because you just get the gear acquisition syndrome, gas as they call it, and next thing you know, you're you're more interested in how this is different from that. And it has nothing to do with actually learning your scales, learning your chords, correct? You know, correct? Practicing

nothing. Yeah. It's just hundreds and hundreds of hours. And most people think, okay, if I can play, you know, a three chord song, that's that's, you know, that that means I've learned. And that's just the beginning. Like there's so many layers to this. And you got to spend 100. It's the same principle with coding. You can have course acquisition syndrome. You can have language acquisition syndrome, framework acquisition syndrome, but ultimately it's all programming is is just basically instructing the computer to do your bidding and they start to have the same pattern after a while. You know, it's just like you can write a book in a great novel can be written in any language. It's the story that matters.

It's how you craft the novel that matters, right? It's the same principle. And I I think we, you know, those of us who stuck with it long enough realized it was never about the the language or the framework or, you know, this course or that course. It was always about getting the right information from the right place that works for you and that's in individual and then sticking with it for a ridiculously long time.

Yes. It's it's just a matter of time and effort right? It is time. Um yeah. Well, I want to go back to, you know, 2018, I guess, your final attempt at learning programming before you broke before you met your evaluation criteria of getting paid to code as an employed developer. uh and let's say hypothetically that you were doing that in 2026 where we have you know cloud code and we have uh GBT 52 and we have like all these different tools that we could basically go in and just learn in just in time and we don't necess like

and and you've got at the same time on you know Twitter on uh you know like Tik Toks showing you like oh you don't need to necessarily do this or do this that like a whole lot of crazy contradictory advice it's extremely noisy as you pointed out

like what would your learning process how would it be different? Yeah. So, so I'll give you an example from one of my mentees, right? She um one of the things I get my mentees to do is I have an entire training on asking good questions, right? Because this is a skill that they need to have at the workplace, right? Uh and asking good questions forces you to understand what it is that you're trying to achieve, what it is that you don't know, what it is that you've tried. Um and it's the same princip. So the same principle applies to well what do I need to learn?

You know that that is not a great question. Uh a better question is what do I need what am I trying to achieve? What do I need to learn today to get me closer to that direction and what do I not need to what do answering the question what do I not need to learn in my view helps a lot with then choosing what you need to learn. So one of my students, she had a problem and she was asking um and she asked good questions and rather than give her the answer directly, I was training her to ask better questions because then she could sleuth and investigate her way to the answer. Yes, it was longer. It was more painful. It took a lot more time for me and it took a lot more time for her, but she was learning critical skills in the process. And then eventually she said, "Oh, I'm just going to ask Chat GPD." Um and she asked chat GPD and she came up with the answer and she sent it to me and I said, "Okay, looks all right. Uh

I'm not going to say anything." And I gave her a prompt, right, Quincy? I gave her the prompt and the prompt is this is the answer you gave me last time to this and this question. Um, now pretend you're a very senior software engineer and tell me why this is wrong.

And chat GPD told her in detail why she was wrong, why its own answer was wrong.

And she came back to me and she said, "This has just blown my mind. I've I don't know what's right anymore." And I said, "This is always the problem is even the expert that you get advice from, you don't know if they could be wrong in your circumstance. Right? And chat GP doesn't know anything. It's calculating the probability of every next word. It doesn't know the answer. It's calculating the probability of words in sequence. Okay? So don't assume it's omniscient. It doesn't know everything. It's just really good at calculating probability. Now, as so happens, its answer was directionally correct, useful, a good stepping stone, not entirely wrong, but it then went and said its answer was wrong. Right? Now, who do you believe? So, and that's why I told I said you have to think about this from first place.

It wants to do whatever you told it to do. So, it'll figure out reasons why what it previously said is invalid because trying to make you happy at the end of the day. That's the entire reinforcement learning with human feedback process is getting it to be more palatable to humans who as you said are shortcut finding machines essentially.

And you have to give it a personality and a role. So if you tell it, you know, you're not afraid to disagree with me, it will actually be a lot less agreeable. It just will, you know, you have to give it the right instructions. And so if you're learning, learning has never been um information has never been the problem in learning, right? Um Quincy, you've seen this as well. It's always been know it's it's it's been knowing. It's to know enough that you know what's worth learning and what's not worth learning and to have an intuition about whether this feels right or wrong. Even in engineering now like it's it's not like I learn to code and I stop learning. I'm constantly learning new things. The big difference now is that I have an intuition when things go wrong about where to look. And it's the same thing if you're asking chat GPT. If you don't know enough to judge its response, the validity and usefulness of

its response, it's no different from a random Google search. It's it's the same thing. It's it's not helpful, you know, to to not know enough to judge the output,

right? And if you were doing a random Google search, you'd have all kinds of other kind of like tells that would help you maybe ascertain whether the information is valid, like do they have a active, you know, SSL certificate on their page? Like when when is the copyright date? Was it updated recently? You know, like all these other things to give you an idea as to like

little cues. Yeah. Yeah. You know, uh but you don't necessarily even get that. You just get the the raw output. So, uh but you can of course look at the links. Hopefully, it'll output some sort of reference. Uh, right. But, but I I definitely see what you're saying. One followup I have for you is okay, let's go ahead and take as a given that LLMs are useful and that they do help you get things done and they help with the ideation process. They help you reason through things. Um, but how would you compare learning through asking and having conversations with LMS to learning from a textbook that was written by a bunch of PhDs and peer-reviewed by a bunch of PhDs and essentially structured in a way to get you from not understanding anything about math or like a specific field of math like algebra to having a reasonable grasp on algebra and being ready to move to the next algebra book.

Yeah, it's a great look great question. Um, When I was in law school, like this is we're talking, you know, almost 30 years ago. It was a ridiculously long time ago. Um, this is before the internet was was mainstream. And one of the things I did learn from one of my professors was, you know, to to understand what the role of a curriculum is. Um, and many many years later, this came back to me after I did my MBA. And then I realized, I don't know, maybe should I have done that? Should I have not done that? But this was critical when it came time to learn to code, right? is if I didn't have a curriculum, I was going to fail, right? And so the role of a good textbook is most people will look at the textbook, look at its title, find out if somebody recommends it, and maybe they'll look at the table of contents, right? The role of a textbook is in my view is actually the inverse is the table of contents is

there to tell you all the things you should not do. It's actually the negative space of the information which is the is this textbook sufficient to get me to my next goal and if I stick to what it says in the TOC will I get to my next my intermediate goal my stepping stone right but the benefit of that is not just the content that's in there it's the fact that in forcing you to focus on that it's eliminating everything else right it's that negative space and I think that's the role of a good textbook is to be good enough that you stick with it end to end learn what you need whether it's one chapter or you know or the whole thing um because there is no shortage of information in the world and the problem with textbooks is not that the information is not right.

The problem with textbooks is you may not stick with it long enough to learn what you need to learn because you will jump to something else

because that's what humans are doing. They're constantly looking for shortcuts and that's what you and I probably both did. I I certainly did when I was learning coding. I was constantly secondg guessing okay is this actually like I'm following this tutorial but like may maybe I go follow this other tutorial. Maybe that'll be easier. You know, the grass is always greener on the other side, right?

100%. And you know, I'll give you an example for music, right? I'm I'm practicing at the moment for a gig that's coming up. Um and and and you play bass and you play guitar, so I'm I'm I've never been a good metal guitarist, okay? I'm really very bad at it, but I'm trying to learn this piece, so I have to do a bit of tapping. And um I have only a Strat. And the problem with the Stratacastaster is when you tap on it, there a lot of that buzz buzz noise comes. I'm thinking, oh, should I go out and pick up one of those, you know, humbucking metal guitars where you could tap a little bit? I'm like, "No, no, this is not about the pickup on the guitar. This is about the fact that my technique sucks."

You know, just practice the acoustic. Yeah. Ang Min, the famous Swedish guitarist, I think, from Sweden. He does all kinds of crazy tapping and he always plays a Strat. Uh that's like

I didn't know that. There you go. Yeah. And he's like one of the neocclassical metal gods, if you will, right? Uh so so it can be done. It's just a question of technique, right?

Technique. And but I was looking for the easy option. Oh, maybe if I just go and drop, you know, a few hundred dollars on a new. That's not the solution. The solution is get good of the tool you have, you know?

Absolutely. And I think a lot of people just want to buy their way to good skills. Like I heard it described this way, like people love going to the bookstore. They love picking out that teach yourself Spanish book. They love putting it on the table and and like checking out because it feels like when you're buying that book, you're also buying the time and the willpower and everything else to get through it. But it's not a question of money. It's a question of

prioritizing and allocating time at the end of the day. Totally. And and that's why going back full circle to the library conversation like you know back in history less so these days but there were people who you know taught themselves the law they taught themselves the basics of economics. Uh just going to the library. And this isn't you know we're talking 200 years ago when there weren't a proliferation of books. There was maybe one book you know and you had to find your way to it. Um Emerson you know taught himself a lot that way. Uh and I know that there are business folks today in who are alive today who came from nothing who spent a vast amount of time in the library and only did the only two books that were in the library that because that's all they had. But that constraint forced them to complete it and then translate that into action. So they didn't keep hopping around wasting

time hopping. They did whatever was available to them and then translated that into action and just kept trying and they just spent ridiculous amounts of time doing that.

Yeah. You're somebody who has changed careers a number of times. Speaking of hopping a lot uh and three times of course you did you did like the JD jurist doctorate like become a lawyer. Uh you did the masters of business administration. You may hear us saying MBA. We are saying MBA and not MBA of National Basketball Association just to get our acronyms uh laid out there for everybody. Um, and then of course you learned programming and went to work as a software engineer and probably went through an extremely arduous process of learning the best practices that they teach at Google, right? So, so you've gone through this kind of gamut, this entire gauntlet of different learning and what are some observations you have about career change that may not have occurred to the person who's still in their first career and trying to decide how to transition into their second?

like what have you learned along the way about the nature of starting over in a new field?

Yeah, it's um oh wow that that's a deep question. So I'm going to try and and and answer it succinctly, but I think most people think that learning to we'll use your example from the from the National Basketball Association, right? Most people think that learning the rules of basketball is enough to to play the game professionally. uh or sorry most people would not think that because they know there's there's one thing to know how to play basketball is another thing to be good enough to to get into the NBA to go professional it's the same thing with any skill there is the literacy right and then there's the level of skill that you're market competitive enough to beat incumbents okay and there's a huge gap between those two um and the gap is often not just knowledge um you could I there are I'm sure there are plenty of junior of career transitioning developers out there who are better than some of the

junior developers with two years experience. I have no doubt that they exist in droves. But the problem is the market cannot take a risk in them. Right? So, and this is something I spent a lot of time with my mentees on is don't just focus on the learning to code. Focus on the extremely difficult part getting noticed for interviews. That's like I have an entire seven stages thing. I think I've you know even talked about it on some of my free code camp blogs when I've shared my journey is learning to code is like stage three.

You have four other stages after that. You have to get get noticed by recruiters. You have to get interviews. You have to perform well in the interviews. Then you need to select the right role because too many people forget that once they're a software engineer in 5 years or so, they're going to want to either be promoted, have a different role, have a different level of responsibility. You're not going to want to write code for the rest of your life all the time. Some people may, but a lot of people will want to grow as part of their career. And if they're not thinking about that, they'll accept a really bad first job, maybe get laid off and then never find something, you know, meaningful or don't know because they don't know they don't have the skill of being able to work in or attract attention in the marketplace. And this is true of startups too. This is the lesson I learned in startup, right? I

built a great technology and I did not market it properly and so it failed, right? It's not enough to have a great product. You have to convince the market that it's better than the alternatives. That's a very hard thing to do. learning the skills actually easier than convincing a marketer to take a risk in you. Every single one of the new careers I had before I asked for a job. When I spoke to the people in that career and I said, "Hey, could someone like me do this?" Everybody was wildly enthusiastic. Yeah, of course. You have the personality, you have the skills, you know, anyone can do anything. All that really good bumper sticker stuff.

All right. Will you give me the job? No. Yeah. Exactly. Right. You just got done telling me that you think I can do it. Why? Why aren't you giving the job? Well, I'll have to defend it to HR. I have other alternatives who are who've got runs in the board, who've got experience in industry. It's a difficult to defend decision. Ultimately, it boils down a hiring risk. Now, I've been a hiring manager long enough in all my careers to know that that's a very real thing because it's it's not often, especially in bigger companies, that hiring is just a oh, I've taken a call. No, it's a process with committees and other people or you know other people involved and your boss involved and your job is to find the best candidate you can in the pool that comes to you that in the pool that's available to you and just because you know like and trust someone may not be enough if there's a is a serious risk

or perceived risk. It may not even be a real risk. It's a perceived risk that this person

there could be some other staff member who's like I'm not sure about this guy. they don't he doesn't have a four-year degree or correct uh look look at how little experience why is there this gap of like three years in a resume or something like that right like there the the perceived risk is something that you have to counter in addition to the actual risk that you can actively

you know you can position yourself in a way to address some of that risk right and ameliorate it with the correct technique

with the correct technique and and the correct amount of time in the market proving the job but ultimately you have to prove prove that you have to derisk yourself at every one of these seven stages that I talk about. You know, you have to d risk yourself when you're asking recruiters for time, right? Because recruiters will have lots of people that they're trying to put, they get paid. A lot of them get, you know, eat what they hunt, so to speak. So, if they put the wrong candidate in front of somebody else, they've lost their commission. Like, it's that's a sales game, too. I've had recruiters in my podcast talk about this. So, you have to understand the recruiter's point of view. Then, you have to understand the interviewing point of view from each of the people you're interviewing with. you know you have to when I was at Google I was very lucky 100% of my interviewers were um were very friendly to

nontraditional backgrounds right but when I was at Google I was also training to when I was doing my training my shadowing to be an interviewer at Google there was one individual who is very old school whose yard stick was if you don't know what I know you're not good enough for the role not actually considered to be ideal but that was that person's point of view and I shadowing them and I was like if I had this person there's no way I would have made it in

because the yard stick is you have to know what I know that's not a great yard stick for interviewing but it was their yard stick for whatever reason right and so so much of that depends on the luck so you have to you know once you've got through the recruiters now you have to have the luck of the draw in the interview as well you know on the day you have to make sure that they're open to you immunable to you and there are ways you can improve the odds of that but you have no way to control that situation and even there they were assessing risk when I was interviewing for Google when I interviewed candidates now a lot of it is How risky is this hire going to be? And it's not necessarily phrased in those words. It's can this person do this job? How much supervision will they need? Will they fit in with the team? Will they be able to handle things autonomously? Will they make good judgment calls? Will they

communicate upwards appropriately such that I, as the manager, I'm duly informed of things I should know about, but not overcommunicate so that I'm saddled with always, you know, having to think for them.

Try to balance. It's it's Yeah, it's hard and it really depends on what our goals are for the next two halves or, you know, the next four quarters, whatever it is. Like there's all this stuff. So people need to if they if career changes underestimate the transition side and completely overindex on the on the obvious bit which is learning the skill whatever it is.

Yeah absolutely and and my mantra is you need to build your skills you need to build your reputation. You need to build your network because those other things, your reputation and your network, help reduce perceived risk so that you'll be more likely to actually be able to get the job where your skills will then carry you.

Yeah. Um, you know, your book on this, so I've listened to it a couple of times now and that was after I made the transition, right? So, it's you narrated it. Um, and it's you talk about all this like 100% of what you said that I was I was just I remember listening once at the gym nodding vehemently to what you're saying cuz I'm like yeah you honestly people should just listen to your book and stop stop indexing for just the learning to code bit which is what everyone's eager to get into because it's the obvious thing. Yeah. It's like everyone wants to buy the guitar without have thinking about the years of practice that comes after that with bleeding fingers, right?

Everyone wants to just look like the rock star without actually doing all the work. It's the same thing and your book covers it all because your book goes deep into all the phases and steps of it. It really does. It's great.

Well, thanks for the endorsement. It's episode number 100 of the podcast if you want to hear me read it with occasional piano in the background my son playing. Uh but yeah, I had a thank thank you. That means a lot to me. And uh I know you've spent a great deal of time managing people and also working as an engineer. So the fact that you're saying those things is incredibly edifying. Thank you. Um, so I am very interested in thinking about the role of management. Now, we've talked about managers as people that can potentially reject your application, but like if people do ultimately want to get into management, as you have, you're an engineering manager. Uh, what are some aspects of managers that you think are really good? Because one of the things you said, I'm going to try to quote you directly here.

You said there's something addictive about watching someone become who they've always knew they could be. That's what I love about managing people.

Yeah. It's it's and it's something I I found out well before I became a software engineer. And you know, it I guess it reflects one's individual philosophy. I think people, you know, the fact that you started free code camp means that you're fundamentally optimistic about what people can do. You believe that people can acquire hard deep skills. Um, you believe that they can transform their lives with it. Right? And you wouldn't have done it if you didn't have that fundamental foundational hypothesis about human nature. And I share the same optimistic view of human beings. I do not believe anybody, no matter what Reddit says, no matter what modern day social media says, which I think just optimizes for the negative and not the positive, and the vast majority of the world is actually positive, but it's not not visible. Um, people want to do good work. They really want to thrive. They want to succeed. They want to feel good

about themselves. They want to feel competent. They want to feel appreciated. Um, and surprisingly, people are actually very good at accepting being corrected. um sometimes even being reprimanded when appropriate when it's fair and it's clearly in their best interest. Right? So, and I was very lucky in that I when I say this, it's going to sound weird, but I was very lucky in that I've had some really horrible managers way back in the start of my career. And you know, um, and I've also after I had some really rough managers, I had some really good managers. That that's why I got lucky is I could see the difference between the two. And and the reason why management is important to me is it's it's actually more about helping people to realize that they could love their work and feel really good about their growth.

Um because all human beings are self-interested. It's ridiculous for managers to think that you know the the employees only about getting the goals of the company done or what it's

it has to be what are your goals Quincy? What are your goals? Right? Not just at this company, what are your goals as a professional, as a human being, as a family, but what are your goals? You know, now let's try and align that with what the company needs. And if you look at it, there is always going to be a way. Not always all the time, but there's always a way to broadly align it enough that people stay engaged and highly motivated. And most people aren't as motivated by money as you think. And if you're hiring those kind of people, you're kind of in trouble anyway. Most people want a fair wage, uh, an understanding and empathetic boss, but most people seem to want growth in my experience, and it's a very limited experience. They want growth. Now, growth can mean different things to different people. And delivering that growth to them constantly as part of the, you know, conversation cycle every

week if if you meet them every week is what makes people actually transform themselves because often they need somebody else to believe in themselves believe before they can believe in themselves. uh and they need to be pushed a little bit and that's I mean I've learned that over the years as a coach as well. Um and then they do things that surprise themselves but they always deep inside knew they could do it. They're just afraid to to believe that until somebody else did. You know that that's what I meant. That was a long drawn out answer. So, sorry,

but it's a helpful kind of rubric for assessing potential managers, like if you're applying where you're talking directly to the person who's going to directly manage you, which a lot of times you do have that luxury when you're doing a job interview, is this the type of person who cares about me be outside of the scope of just getting my work done at work? Is this somebody who is going to find opportunities for me to do work that will help my overall arc of my career

irrespective of like like it's helping the company. Obviously, it's worth doing within the company, but is this work that is particularly well suited for me given my long-term goals? And managers who make the time to understand what your longer term goals and don't just pretend oh

you know, like the old the old joke is if somebody asks you uh you know, where do you see yourself in five years? You're supposed to answer working for you, you know, but the practical reality is like, you know, where do you see yourself 20 years from now? I mean it it would not carry water to be able to say uh you know oh uh working for you right that that would people probably wouldn't that would

but say oh yeah like like oh I'd really like to be you know in you know manage it at this field or I'd like to be an extremely high level individual contributor right I I talked to uh Santos Yatv a couple weeks ago and he is really advanced in his career just working as a dev he doesn't manage anybody he stepped down from a managerial position because he preferred to just be a contributor right?

Yeah. 100%. And and and you have to know that and your managers to support that and and you know, coach you into that. And it's also kind of disingenuous when um you know I my managers plenty of them have moved on to other better roles, right? So it's disingenuous for a manager to say, "Oh, you know, are you going to be here with me in 5 years? Are you going to be here in 5 years, Mr. Manager? Like you know, or miss manager like you know, are you you know, can you give me that promise?" So I think that's that's disingenuous for anyone to be saying that and all you can say is look if everything works out hopefully we'll still work together and grow grow into things together right um but you even if that doesn't happen and it may have nothing to do with our fault you have everyone has a vision of what they want for their life professionally uh you know buying a home or whatever it is or you know you know having a little bit of

time to spend with kids or you know whatever it is they have a vision and now you work on it together and it's not always going to be possible uh but most people also want to get really really good and they want to know that their careers as risky as the environment allows. Now, 30 years ago, careers were less risky. These days, they are more risky. Not just for us, for everybody, right? The world has changed, which means it's not just about what you know, it's also about being market competitive. It's also knowing about how you build relationships with people, prove your value before you're actually in the job.

There are all these other skills that are required, you know, and even software engineering is changing so much now. Yeah.

And and I think a lot of people push back like, I just want to be an engineer. I went to school for engineering. Why can't I get a job as an engineer? like why do I have to learn all this meta knowledge just to be able to navigate this Rube Goldberg machine that is the the job application process you know so that I can eventually get paid to do what it is that I've studied to do right and and uh one of the questions I have for you is you know 2008 post 2008 financial crisis to really around the release of GPT could be one uh of the like chat GBT setting the the world on fire with AI ambitions and uh of course like zero interest uh like like near zero interest rate

interest money was almost free you know like like these things like they kind of happened at the same time and then that's and also just like employee comp companies deciding like hey we can lay off a whole lot of people Elon Musk just laid off practically everybody at Twitter we can do it and we can get away with it right and just like letting loose tons of developers after like the

I guess pandemic overhiring uh people the companies were trying to hoard the developers so so there was that period between when that happened in, you know, 2022, 2023, um, and and 2008, right? It it was probably a good 15 years or so. And that was

Yeah. You've described it as kind of an aberration. That's not normal. A normal period. We were in like this incredible period of stability

for tech especially. Yeah. Yeah. For deck especially. Yeah. And you talk about like

Yeah. Yeah. What that like the difference between then and now and the difference like how people can adapt if they are still in the mindset of the

totally before the chaotic period of now. Yeah. H happy to sort of sh at least you know I did a podcast episode but it was so boring and dry. So I I you know I don't blame anyone for not not really sticking with it. But but the the bottom line is somewhere in the 2000s startups became a real thing. Okay. it became the new frontier for what human beings could do. Um then the the mobile phone revolution happened you know 2005 67 onwards right like before that it was all blackberries for those of us who remember that right it was only when the iPhone came out that real what you know web 2.0 I know seen as it's called like pale mobile.

Yeah. And to an extent that's when every normie air quotes normie. I've never used that word before but that's what I hear people saying. That's when those average average people started to really get on the internet because suddenly they had like a lot of people didn't have a laptop or they didn't want to spend all day in uh you know a room looking out the window on a browser. Yeah. Right. Correct. So the real explosion happened with mobile computing um you know in that sense and the mobile internet as well and really the mobile int coincided with the global financial crisis wasn't connected in any way just timing wise it happened around the same time right 2007 to 2010 and then the whole world got onto the internet on the phone and then that's completely changed civilization so by the time 2010 came around we were leaving the financial crisis and I was working during that entire time hanging on by my fingernails

it was rough Man, it was so bad. Like people today who complain about stuff about how unstable the job market is. I'm like, you have no idea. But I'm lucky that I went through that and I think you were around then too. It is a very difficult world. Um but anyway, so you you survive that and then you realize how job markets work. Now that prompted me to study the history of job markets a little bit and you know when I decided to be a lawyer, it is because in the 80s and you know I didn't decide in the 80s but in the 80s law was considered to be the next big highly lucrative profession. Come the 90s I graduate school a lot of people are thinking about law because hey it's the next big thing right it went through that curve by the time 2005 came up the number of universities who was churning out lawyers into the market was substantially more than the market could really absorb and that happens

everywhere right there's a steep demand because it's lucrative and then the supply catches up from traditional institutions and then they intersect somewhere and then you have a mature market it's the same thing in any industry automobiles um you know MBAs were really big in the 50s and 60s

then they sort of peted it out you know because financial management was big in 50s and 60s then finance became really big in the 70s and 80s ' 80s especially was big for finance like you know that's where the Gordon gecko sort of period of the world right

classic movie Wall Street which everybody should watch it's never been more timely than I love that movie

it's such a classic um and then you know you had sort of the legal eagle type of movies that came out as well because these were the big professions of the day right they were prestigious. Uh there was money involved. There was power and glamour involved. In 2007 onwards, it was startups, right? It was startups that raising funding. That was the big thing. And so, everyone started to move into that. And then we had that glory 15-year period where it was postrecession. Therefore, the market could only go up. One of the biggest bull runs in history. And 15 years is a long time, right, Quincy? 15 years is enough time for someone to finish college, join the labor force, and then become a manager of managers like up to director level in 15 years. It's it's doable, right? Especially if you're sitting on the nose of a rocket ship. If you start a company when it's 20 people and you started there in 15 years, if

it's giant and listed, you're going to be very senior, right? Yes. So the entire careers were built in a nonrecessionary environment with almost 0% interest rate where there was a huge demand for software engineers. Now that supply is starting to catch up and all industries go to the maturing scurve. Now that tech is starting to mature a bit maybe AI will unlock a whole new thing. I think it it may a new S-curve but when things start to mature when demand and supply starts to meet the kind of options in the job market changes. Now suddenly it goes from being an employees market to an employers market because there's a rich pool of highly experienced people that have enough mobility in the job list and it becomes harder for junior. It was it was easier for juniors in 2013 than it is today. And that's true of every industry in its era as it goes through the scurve. Once you understand those

principles, you then prepare your strategy for which part of the scurve you're in. So I'm very optimistic for the next bull run. But now's the time to prepare and it's going to take two years to prepare for that. But if you get your timing right, you then ride the next massive wave, right?

Yeah. Yeah. And it's better to be like paddling out toward where you think the wave is going to be

before the wave is I mean the wave can still be really far off. You there could be no sign that the wave is even there, but you want to be in position when the wave does hit. Otherwise, you're going to miss it.

And and history will tell you there's always a wave coming. Yeah. Yeah. And and historically there always has been. And a lot of people maybe argue this is the last wave. you know, you're gonna be perfectly part of the underclass working for some guy who caught the wave when you could be the guy who caught the wave and have like all the optionality, right? Um, totally 100%.

I I think a lot of that comes back down to um just people fundamentally like this is it like high school is the end of my life, you know, if things don't work here, there's no tomorrow, you know, like like that. Humans are kind of wired to to not think long term enough and and to to hyperfixate on the present.

Like I certainly am often guilty of that even as I pride myself on being a relative long-term planner.

Um yeah, I I still catch myself doing that because I think it's nature. Yeah. And it's also part of our conditioning like we all there's a reason why the children of sports people do sports or the children of you know engineers do uh do tend to go into more engineering profession or the children of doctors tend to become do there's a reason for that is because you're steeped in that environment. It's the world you come to know asically even if it's not your formal education it's the world you're in. You see children of business folks know business way more than the average person does. Not because they're naturally smarter just the environment they've come from right like that's how they've seen the world.

They've observed things at the dinner table on weekends and stuff like and you learn all this. It's the same thing with understanding careers. Most people think that our parents went to college, got a good job, held on to it for a very long period of time, moved maybe twice in their entire career, but there was a stable environment. There weren't a ton of new skills to learn all the time. And so they think that's how their future is going to be. But the world around you is changing. You're in the change. And you know history is not what it kind of used to be in that sense that you can't use the past to reliably predict the future unless going back to what you said earlier you know you you focus on some of the meta principles around let's say learning you know that a good textbook will be good in any environment the question is are you picking the textbook that's directly aligned to your goal

right you know every textbook is going to be great for somebody if it's directly aligned to their goal

right meta principles yeah I I I often just say pre-ooking is essentially an interactive textbook for software engineering,

right? And and and that's kind of how I think of it. Uh so 100%. So man, we've covered a lot here. Uh one thing that I do want to ask you is you've said that like LLM tools are changing a lot of how people think about skill in general. uh and that now kind of meta-kills are coming to the four uh non-digital skills is what you called them. What do you mean by non-digital skills? I mean, you talked a little bit about like being able to empathize with the hiring process holistically instead of just saying, "Okay, I'm going to talk to this person, then this person, then this person, then I'm going to find out whether I got a job." Uh but instead actually understanding like how they're trying to minimize risk and stuff like that, right? Okay, that could be considered a skill like understanding organizational behavior, but like what do you mean by non-digital skills in general?

Um, so you know in the there there's skills sort of in the workplace, there's skills for life. I'll talk about workplace skills. Um, as long as as long as we're working with other people, being able to do that well is an essential essential skill. Now, there's one big difference. I remember talking about this with John Cricut, another colleague of mine in the UK, uh, on his podcast, right? He's he's a very big LinkedIn presence and stuff like the the reason why compilers are generally easier to work with is that they will tell you upfront when they're wrong.

They'll give you immediate feedback. They will complain straight away. Human beings don't do that, right? Compilers are highly deterministic. You and I do the same wrong thing, it'll give us the same wrong message, right? But you and I say the same wrong thing to different people, we won't get the same response. reliably. It depend it'll depend on our report. It'll depend on context. And human beings can sometimes just be quiet about things and not say things, right? And so

are you talking to a Californian or are you talking to a New Yorker, you know? So that is true. Part of the stereotype, but but yeah, like New Yorkers are generally going to tell you you're wrong without or like Germans will be very blunt in my experience,

very direct about it. Yeah, 100%. you know, and the same thing here in Australia, like Australians are very, you know, quite culturally similar to the British in that they're very circumspect. They're very polite and it's or Japan, it's so much is not said. So much is not said, but you know, if you're part of the culture, you'll read the signs. That is hard, man. That is so hard compared to writing code because hey, the code the compiler will always tell you. It doesn't matter, right? Um and so that when working with other people, being able to work with other people meaningfully, have the right influence, have the right credibility, these are skills, uh being able to communicate, uh is I know people get kind of annoyed about it, but you know, the best lesson I ever learned on communication was when I came very close to losing a case actually. um it didn't end up going to judge trial anyway but

um you know I was in a small court in India and I was arguing something um and it was a court where you know everyone typically spoke in the local language or were bilingual and I was there with my fancy legal education from a top tier law school using big words cuz you know I I I read you know fancy English books. Um and then after the thing the judge pulled me aside and he said look I'm sure you're very good and you're very you're clearly very smart. There's one problem I didn't understand anything you said right? Um and he taught me the best lesson ever. He said the purpose of communication is not to say great things. The purpose of communication is to be understood.

Right? It is a completely different frame changed my life. Quincy that one lesson I use it in interviews. I use it in managing I use it in crossunctional sort of collaboration at work. Right? Be understood. Be understood. Be to be effective. Right? It's kind of what we do with code as well. If the compiler doesn't understand us, it's going to tell us. But if human beings don't understand us, it's not a lot of them won't tell you. Especially when you're working across country like culturally or country. Now I work 100% remote. I I think I work my team's like across six time zones globally. It's hard. It has nothing to do with programming at that point of time. It has to do with all your non-digital skills in that context.

Especially if you're doing remote work or work in larger organizations or really fast-paced work um where you're on a small startup and I've been in that environment too and you've got to prioritize. Now, how do you influence priorities? It's not by writing more code because that's going to take a ton of time, you know. How do you influence direction? How do you influence priorities? How do you choose who, you know, who does what, especially if you're leading things? It's hard, you know. Uh, and that's why I think that's going to be the future. Also, when you think about it, anytime technology makes things easier, the value of that thing sort of drops a little bit on the stack and there are other things that come.

I'll give you an example. There was a time when getting a book published was ridiculously hard. So there were fewer books in the world. So you didn't need to market it as much. Now with self-publishing and podcasts, you can have plenty of really great books out there that no one hears about. So what are the skills now? You have to get on other podcasts. You have to really market it. You go do the podcast circuit is really big for book publishers. Why do they all do it? Why did uh you know uh why do politicians these days for example get more on TikTok or get onto podcasts? Because the way you market your skills has changed. And it's the same thing in software engineering when when LLMs can do a lot of the order complete or the codegen for you. It's the design work, it's the judgment, it's the taste, it's the analysis, it's the review because we're all even IC's today are going to be managing they're going

to be managing LLMs, right? You are going to be managing an LLM. You are going to be managing a team of LLMs probably at least for 10% of your work something. And that's down to communication. What is a spec? What is a good prompt? It's communication. You know, it is helping the LLM understand your goals. And just like a human being, you can't expect it to be a mind reader. You can't expect the LLM to be a mind reader as well. And you have to know, you have to understand the LLM to be able to prompt it. So telling the LLM, tell me why you're wrong means you understand how LLM's produced the next token, right? It's the same thing as understanding a human being. How does Quincy do his best work? How does Zuben do his best work? What does Zuben not like? Or what does Zuben not respond well to? What does Zuben not do well?

Right? You have to work, my manager has to work around those things, too. I have to work around that with my team and understand each person. That's hard because it's completely non-deterministic.

Yeah. And again, because you're not necessarily going to have a tight feedback loop like you'd have with a compiler. I love what you said. The purpose of communication is to be understood.

You also said anytime technology makes technical things easier, then less technical things become harder. Writing a getting a book published

used to be really hard. Now it's just a bunch of like emailing files back and forth and and maybe like emailing back and forth with a publisher assuming you have a publisher or self-publishing and then it's the marketing. It's the distribution that be becomes hard.

Yeah, it gets really hard. Yeah. Yeah. So, having recapitulated that, this is the first podcast of 2026. I didn't mention that, but like welcome back. Happy New Year everybody. Um, I want to know what are your big goals for 2026? What do you hope to accomplish this year?

Yeah, thank well um I think I'm at a stage in my life when uh you know I I I want to be uh I I I've realized now after 20 years 20 something years of being in the workforce 25 years almost that what what matters to me most most is is having the freedom and the control to learn the things I want to learn. Um, I really enjoy the work I do, the company I work at, the people I work at, I love it. I I want to continue to grow and be of service to them. Um, you know, I I want my team to succeed. So, that on the professional front, that's what I want to do. Um, on the more personal front, I want to spend more time playing music. And I definitely want to keep I've been teaching myself the LLM, you know, world now for about a year. It's changing so fast. So I think what I really value there is that I I had when I learned to code the first time around develop the ability to learn like learning how to learn is a real meta

skill. Um and I'm looking forward to stress testing that because honestly being an engineer now is quite different from seven years ago when I started it really start and not just because of the management role like just the actual act of coding is quite different and there are ton of skills involved there and none of them are you know best practices haven't evolved yet. So, I think being able to influence the way I work with these tools, being able to help others in my team work better with them is something I look forward to learning because I think it's going to be the next iteration of what software engineering is all about. Um, you know, and and I I'm really excited to learn that because it's it's such a privilege to be in an environment where things move so fast that you have to learn fast. It's great. It's stressful but great.

Awesome. Yeah. So, it sounds like you by the way. Sorry. Because I I know you're going from stress and strength. You guys are doing non non-coding curriculum. You're expanding this. What do what what do you want to achieve for 26?

Well, um we're hard at work on finishing version 10 of the coding curriculum. So, free code camp, like we're still mainly focused on programming. That's where like a vast majority of our resources go is to the core interactive curriculum. Uh but we are going to have we already have uh a little bit of the Spanish curriculum and of the Mandarin Chinese curriculum. And of course, we've got a lot of English as a second language coursework. we call it English for developers. So that's already live. Um and we are going to continue to work on that. That is free cooking's main priority this year is just the same thing has been really for the first 11 years. It's going to continue to be for year 12, year, 13, year 14. Correct. Is

just heads down creating interactive coursework and teaching people deep hard to achieve skills, giving people the textbook that they need to be able to get in and get out with confidence and not languish for years. like uh I mean I can imagine if you just try to teach yourself Chinese just having conversations with elms and you didn't have a book of grammar you didn't have a book of like pronunciation you didn't understand like I've learned Chinese for 22 years

that's my big personal goal this year is I've been reading a lot of Chinese like I'm talking about like several hours a day like just reading Chinese trying to really firm up my understanding of the Chinese characters and uh so that when I go back to China last year I went to this awesome conference in Beijing being uh put on by the Apache Foundation uh community over code and I want to go back and I want to understand way more than I understood the first time. I did have a great time there. I gave my presentation partially in Chinese. I had to switch out. It was just taking too long. Yeah. But but uh I I'd love to give a presentation fully in Chinese because my Chinese is a lot better. And I personally believe that just reading comprehension and just massive exposure to a whole lot of text is going to help me get there. Uh so so that's what

I really hope people are listening because exactly what you said there a few hours a day massive exposure you know a lot of repetition and it's not really about the vocabulary it's more about the immersion into the subject and not just acquiring new words I really hope people listen to you about that because that's really what it takes to learn anything it's immersive it's not you know with as much of just persistent every day showing up and just cranking it out without an expectation of a fixed timeline you know yeah I really hope people listen to what you're saying there it's so important Yeah, I mean my my philosophy has always been get the fundamentals in place. Make sure you have like near perfect pronunciation. Make sure you can reproduce all the sounds of the language. Yeah. Make sure you understand like the the stroke order and radicals and you know all these different considerations you have to do with

Chinese that you don't necessarily have to do with Spanish. But uh then like once you've locked in on the fundamentals, then just go volume.

Yeah. And I I would say the same thing for people that after you completed the free codeam curriculum, just go build a ton of projects. You're already going to build a ton in the free cooking of curriculum, but build even more. You know, spend thousands of hours building additional things. You're going to get so much better. Just like the saying quantity has equality all its own.

I certainly feel that's true for learning. Hey, if it if it works for LLMs, why won't it work for us? Like they literally use quantity learn, right? And see the

pattern. I mean they read the entire internet and it makes them way better at answering questions that they were just training on like the brown corpus or some of these different like model like uh corpuses that were corpai that were available back in the day like when I was correct when I was getting into programming we learned about uh natural language programming and I learned how to use the NLTK the Python library natural language

I remember that. Yeah. Yeah. and and like it was comical the capabilities of this. Even if you look at like GPT2, I remember using GPT2 a lot and just like how I was like this is a fun toy and then I'd forget about it and go and do stuff like it it really took a lot of volume. It took a certain degree of quality before it really became useful. U and I think the same thing happens with skills like

and it's the same principle as reinforcement learning. Yeah. Yeah. You know, you just keep doing the loop, get the feedback, improve it takes volume. Yeah.

Yeah. And I guess that's kind of like what I'm doing with my Chinese with your guitar.

Well, I'm trying to because I neglected the guitar for so long because I didn't know how to learn, you know, and I um I just thought, oh, if I do a little bit here and there, you know, and and then I was wondering why I was plateauing and I realized I was plateauing is now in hindsight, I realized I was plateauing because I wasn't intentionally trying to improve a specific small thing, you know, and it comes down to technique after a while. um you know so yeah it's small things like once you understand how learning works you realize how much time you lost cuz you did it wrong cuz and what made me think about this just as a little aside which may be interesting for some of your listeners is you know the the um uh the Bob Dylan movie um what's it called um No Direction Home or something you know the one with Tim Timothy Sham

yeah well both him and um the you know um the actress Monica B I can't remember last name who who plays is um drone bays. Both of them learned music and guitar from scratch in four years to prepare for that role and they did all the songs themselves. And I'm talking about advanced like folk style strumming, right? I looked at that and I'm like that is incredible. That means with the right technique you can can get so good at something, right? And you just have to get the technique right. And so I'm like I should be doing that with my guitar too. Like I've done it with code, so you know, and I've been playing guitar for 20 years. I've not been learning for 20 years. I've been playing for 20 years. Now I'm starting to learn the guitar, you know.

Yeah. Yeah. Yeah. Locking in getting those fundamentals and and being intentional about how you learn and not just totally trusting volume like basically of what I was just saying. I'm just doing tons of volume right now. But what preceded that was a whole lot of directional

oriented, you know, learning to set me up to to okay, I'm like in the kind of like I've beaten all the campaign mode and now I'm in the open world mode. I'm just going and doing all the side quests now. That's kind of how I feel about Chinese. Now, not that's not to say that I'm like native level or anything like that, but I am very proud of the proficiency I've attained and and and I do believe that like once you've done enough of that directed learning, you should be able to just go, "Okay, cool. I'm just going to go learn specific songs and maybe you don't need the same directionality, but there's a there's a chance that maybe continuing to have that kind of intentionality to your practice will continue to unlock marginal benefit."

Yeah. And it also trains you to not just look for the sugary highs, right? Like to just go and jam to one riff. Like it trains you to think, okay, part of my development is drilling the fundamentals. It's kind of much less sexy and fun. Um, but it is so important to unlock everything else just doing some drills, you know, regularly. Yeah,

absolutely. Well, don't neglect those fundamentals, everybody. uh Zumin Pr it's been an absolute pleasure having you back on the podcast uh after seven years and uh I would encourage anybody uh who wants to learn more about his backstory because he did give us like the 30 second you know very quick version uh if you want a time capsule if you want to hear you know my rate of speech and everything like I feel like I I listened to it earlier today to prepare for this and I was like man I talk so slow um I part of it was we wanted to make it more accessible to non-native English speakers who were listening, but you know, now there's like captions and stuff like that.

Captioning. Yeah. And I think back then in 2019 or whenever we did the episode the first time, I think you were restarting the podcast after a long time. There's no video. Like you know, you can see people can see the evolution of Free Code Camp of Quincy. Like there's so much that's changed and you know there's so much growth that's happening. So it's great. So thank you for having me back on. It's always a pleasure to talk to you. One of my fondest memories is having dinner with you here in Melbourne, you know, several years ago. um you know we and then being there at the the um the San Francisco this is when I was learning to code by the way I hadn't really gotten any good but you had that at the GitHub headquarters in San Francisco yeah the the J that's it hackathon

the hackathon yeah you know we caught up there and like yeah it's been such a pleasure to be you know involved in your orbit in in in whatever way I can to be able to contribute to free code camp I've always been a fan of the amazing amounts of service you've done for the community uh so thank you again for having me on and as always Lovely to talk to you, man. It's great.

Yeah. And thanks for everything you're doing to mentor uh folks and help them get into software development. Uh it's it's a lot of work coaching people. Uh it's not an easy field to get into. Everybody tuning in. Uh, until next week, happy coding.

Questions about this episode

What is discussed in "First developer job at age 38 with lawyer turned software engineer Zubin Pratap [Podcast #203]"?

In order: 0:00 Free Code Camp, Google LLC, Algo Monster, Artemis mission; 10:00 Free Code Camp, Google LLC; 20:00 Free Code Camp, Google LLC; 30:00 Google LLC, Free Code Camp; 40:00 Free Code Camp; 1:00:00 Free Code Camp; 1:10:00 Free Code Camp, GitHub, Zumin Pr.

What is mentioned in this freeCodeCamp.org episode?

10 people, works and ideas are mentioned, including Free Code Camp, Google LLC, Quincy Larson, Zuben Proup, Johnson Space Center, Artemis mission. Each one on this page links to its own page with every episode that mentions it.

How long is "First developer job at age 38 with lawyer turned software engineer Zubin Pratap [Podcast #203]"?

The episode runs 1 h 13 min; the transcript is about 15,602 words in 163 timestamped sections.

More from freeCodeCamp.org

About this transcript: the episode's spoken words as captioned on YouTube, split into timestamped sections. Reclip.Me locates every person, work and idea in the text, keeps the time it is first named, and quotes the words around it verbatim. Last updated January 24, 2026.