What Does the AI PM Job Market Actually Look Like Right Now?

Watch the Episode

Transcript

Shobhit: Ready to move to the next level in your product career? I'm Shobhit from Intentional Product Manager. Join me as we discuss ways to help you stand out in your job search and your career so you can have more impact and make more money. Aakash, welcome to our podcast, so great to have you here.

Aakash: I'm a big fan of your work, so I'm excited to be here and glad we can chat in a little depth.

Shobhit: Awesome. So look, probably most of the people who are going to be watching or listening already know a bit about you, but I would love for you to tell us your product management and founder story — what happened, how did you get into product, what your career was like, and then this awesome transition that you've made. Would just love to hear it from your own mouth.

Aakash: Sure, I can give you the short version, let me know where you want me to go deeper. So as a young youth I was really lucky that my dad was a professor of computer science, so I was enrolled in programming classes from sixth grade itself. By ninth grade I was operating an eBay store that I ran my own website on in order to save some fees from eBay, I built my own computer. In college I was working with a variety of startups in and around the university I was at, and it was really full-stack product management — coding, designing, shipping to production, but spending a lot of time with customers, spending a lot of time with executives saying "this is why we should build" what I've now built. That led me into my first product management job at a B2B SaaS company called ScoutForce, which sold into 40% of the D1 schools — University of Alabama football team, University of Michigan football team, some of the big money programs. It was software for their recruiting teams — they have all these field recruiters, and I used to go meet with them while they're doing their thing at football games. Back then they had like 2012 iPhones, they're uploading their notes, they need a good way to put those notes and their videos into the system. So we built that software, and that led me down a variety of different product jobs over the years. I was a VP of product at Apollo.io most recently, and that was actually kind of my dream job for most of my product career — working at a unicorn, being a VP of product, that's what I wanted to do, or so I thought. While I was doing that, I loved it as much as I thought, but I had kind of a second job, which was I had already built up my internet presence to such a level that I had this juncture — like, do I want to, which one do I want to do, because doing both of them, I was probably doing them at 90%. A lot of people say if you do two things, do them at 80 — I was doing them at 90. I was doing 35 hours pretty much for social media stuff, like five hours at night, I barely slept. My stint would be I'd put my kids to sleep and then I'd do 9, 9:30pm to 2am, and then I'd wake up at 7am the next day — sometimes my son or my daughter would wake me up at 6am or 5:30am, as anyone with young kids would know. So I was living that true dual life almost. For me it went down into, well, what is giving me the most passion, what would I do even if nobody was paying me, what would I do just if I had all this time to focus on one — which one would I prefer to focus on, and then which one has the higher upside. At that point I was a VP of product at a unicorn company, and I could see a pretty clear path for this creator journey to be earning significantly more than I was at the top of the pyramid already, so that was a huge factor. And then loving this work, loving writing content, loving talking to PMs and researching PM things — that was a second factor. And then the third factor being my own boss. Right after I worked at that small startup in Ann Arbor, I actually started my own company, hired seven people, worked for four years on an iOS app called Rap2Beats. So I've always had that entrepreneurial streak, like many other product managers. Putting all those factors together, I had to go out on my own again and give it a shot.

Shobhit: That's amazing, I mean absolutely amazing journey. If you don't mind, I'm going to go deeper into some of these, and how you thought through the transition and everything around that. So let's go back — you mentioned your first product role was right after undergrad, that you jumped in — was there a career even before that?

Aakash: That was my first job, at a company called ScoutForce. I was lucky, like most people — I think finding a product manager job is pretty difficult, especially right out of undergrad, usually those people go for an associate product manager. What I did though is I targeted a company with only 15 employees, and I didn't just do product management — I worked with them while I was in college. So while I was in college I helped them ship tons of features, I did the user research, designed and shipped them all on my own. Having done that, they were more than happy to hire me in almost any capacity, and for my first few jobs they used to give me really nice references. They just loved me — I was kind of like another co-founder in that sense, I would do anything for the company because I was young at the time as well. That's how I secured that job — I was working with them in undergrad, they were like, it's a no-brainer, we have to bring you into the team.

Shobhit: I love that. And then did you very intentionally choose product management as sort of the area you went into — how did that happen?

Aakash: No, actually — if we think about what I wanted to do, it was actually to found a huge company. What I looked forward to was specifically people like Larry Page, Sergey Brin, Steve Jobs — those types of people really motivated me. Later on Elon Musk, before everybody started having a million reasons to hate him — these people really inspired me. That's why I worked with that startup from the beginning, I just got handed the title product manager because that was the closest thing to what I was doing. But after that, the first job I actually took was growth manager at a company called ThredUp, which at the time was a Series C company in San Francisco — Series C is an amazing spot to be, because you still get time with the CEO, the CTO, the president, the CMO. I actually worked with those guys on growth, and it was a series where we could work on whatever was most impactful. So for my first year I scaled up a couple things — I took their Facebook spend from something like $300 a month to $30,000 a month at better ROIs than it was at when it was at $300 a month. I scaled up their SEM, using some of these skills that you now see in me today — I wrote all the Facebook ads. I remember there was one ad that went super viral that had a typo, and whenever we'd hire a new brand person they'd always be like, "but how come we're spending so much money on this ad?" And we're like, guys, this is just how it works — this ad has 70,000 likes, everybody loves it, Facebook is serving it at a 60% CPM. So we would fund these types of posts that I would write, and that was like the first leg of growth. But as we scaled our new customer acquisition, we realized, whoa, the LTV on these people also needs to go up, and that's how I switched into growth product, which is all about how do we take those first-time visitors and convert them to first-time purchasers. As I built that, I scaled into the pricing system, because I realized what's the number one thing that drives conversion — everybody knows it, right, price. If you reduce the price you're going to get way more conversion. So I slowly made my way into core product, and then I actually started becoming a core product leader.

Shobhit: That's awesome — a funny story to share on the one you wrote, you mentioned about Facebook ads, one of my best-performing ads has this AI-generated image, and the image has the person with two monitors, and the monitors are looking at each other, it's all weird, and there's multiple comments there, which is like — the first thing you need to do is hire a real designer, and you know that's how you become intentional, this is clearly not intentional. And I'm like, cool, look at the results on the other side. So what works might not be that intuitive.

Aakash: Yeah, and at least at the time — back when I was at ThredUp, we're talking now about 2012 for Facebook ads, the algorithm, it would just find something that it could reliably show to new people that people wouldn't hate. I think back then the ads weren't as good as they are now from everybody else too, and so that post looked organic — most people would comment like, I didn't even realize this was an ad, because we used this image that was blurry, it looked like — I believe I actually took it on my iPhone while I was at our warehouse, so I just took a picture of the racks, it was like a fourth-generation iPhone, so it wasn't a very good photo. I think that's what I tell people who spend a lot on social media — you have to let those people who are working in the social media space cook, you have to let them test things out, it might not be exactly to your brand guidelines, but sometimes those things will just pick up. The 60-cent CPM at the time — the CPMs for the types of ad groups we were targeting was $8-9, so it was like seven, eight percent of the average CPM because it liked it so much.

Shobhit: That's amazing, that's totally amazing. Then tell us more about your product journey — what happened after ThredUp?

Aakash: So at ThredUp I ended up leading a bunch of core product teams and reporting to our head of product, and that set the journey off for the rest of my career, where I was doing basically that — reporting to the head of product — at places like Epic Games, at places like Affirm, and then at Apollo. Those roles were kind of the player-coach role, where I'd usually have one product team where I have to do the IC work for, but then like two or three people at the beginning, up to like seven or eight people that I'm managing and also coaching. Those roles were really great — each role that I took led me to expand my aperture of how I think about things. From ThredUp, which is like a two, three billion dollar market cap, to Epic Games, $30 billion market cap, to Affirm, which at the time its market cap varied wildly but was also around $30 billion market cap when it was peaking, and then finally to Apollo, at a unicorn, which funny enough moved much faster than any of the other companies I'd ever worked at.

Shobhit: You mentioned player-coach, and I find that a lot of people are in that stage — tell me more, how did you make that work, like with a core focus for yourself and then combining that with coaching other people? Any principles or practices that made you successful in those sort of roles?

Aakash: Yeah, one of the things I have a really strong belief about that I think not enough people pay attention to is that you need to be advancing what state-of-the-art looks like for PM work products. That's why I actually often defended the player-coach role, even when I was having like four or five PMs on my team, having directors on my team who had people reporting to them — I still wanted to keep that closeness, because ultimately it's kind of like founder mode, the current thing about founders — it's the same thing when you're a product leader, you need to set the bar higher than anybody else, you need to show them, this is really what a PRD could look like, and a good PRD is not very long, but it is really well written, and the insights you're getting from every line, it's going just deep enough as to not dictate everything to the designers and engineers, but still have thought out everything a PM should be thinking through. It's basically impossible to teach some of these skills — I think going from a nine to a ten in this discipline, it takes like five years of working in product management at successful companies to actually move up to that level. So I firmly believe in actually showing teams, and also the type of impact — what I find is that a lot of PMs build their career on safer bets, I actually had the strategy of taking lots of big bets, 70% of which would fail, but 30% of which would change the course of the company. So I mentioned the pricing system at ThredUp — this system ultimately redid our entire pricing, improved our inventory turns, improved our first new user conversion rate, improved our lifetime value — it was the basis of the business, now they have multiple teams focused just on that, and I just shipped it as one of my many big bets in a quarter. It was a strategy also. And then how did I manage this? I actually think it's very hard — I don't think anybody who wants to work a 40-hour-a-week job can possibly do this, you have to be able to go totally crazy in your job. My typical day would basically look like 8 to 9 catching up on emails, 9 to 5 being triple-booked in meetings, and then sometime in the evening, maybe like 8 to 10 o'clock, doing some IC work where you can show it off. So how can you manage a whole team in just say two hours of IC work a day? Number one, it was really learning how to delegate to my design and engineering partners on those teams, and also picking and choosing teams that could handle that — if you have a very junior designer, you're totally screwed on this front, but if you have a very senior designer who can take a problem statement, go talk to users, create a design, test the prototype, get all the QA edge cases ready for engineering, then you can rely on them. A lot of times I would pair up the most senior product designer with an engineering manager who is product-focused — there are generally two or three types of engineering managers, but the product-focused ones are the ones leaving a ton of comments in PRDs, leaving a ton of comments in strategies and product reviews, asking questions at product all-hands.

Shobhit: That is super insightful, because this is probably the most common question I get, around, hey, I'm just really busy, I can't figure out even time for thinking through strategy and others because I'm so busy in execution — and it seems your secret has been, how do I find the right people who actually want to be more product-focused, who I can delegate these things to.

Aakash: Yeah, there's a certain type of engineer who's like, "you need to write the tickets for me, your PRD did not cover all these use cases," and then there's a certain type of engineering manager who just throws that back in the PM's face. There's another type of engineering manager who's like, "okay, we're going to have a tech lead on the team, the tech lead's job is going to be to embed with the product manager so they understand the questions the product manager was thinking about, so they can answer these detailed implementation results." When you have those types of people in place, when you have those types of teams that are autonomous — a really great example of this is those teams that didn't even need a PM, you'll often find those, like the billing platform team or the payments platform team, but they'll consistently be shipping stuff that's improving business metrics. Those can be very interesting if there are some big bets where you can create some innovation in their space to take on.

Shobhit: That's amazing. Okay, so we talked about the player-coach role, and now I'm really curious about transitioning into what you're doing now — like where did that idea come from in the first place?

Aakash: Yeah, it is serendipity really, or "a surface area of luck increases as you write online," in my case somebody really amazing, Paul Aharan, really pushed me to create a paid subscription to my newsletter. My newsletter was growing very fast just on the basis of writing I was doing out of passion. A lot of people are surprised that I could write so much while having a full-time job, but it kind of just goes back to my whole life — I'm used to working like 14 to 16 hours a day, the average person is not used to working that much, they work eight hours a day, so they look at someone who's working 14, 16 hours a day and think that person must be slacking off in their job. No, actually I just found time for content when you all were watching Netflix. From like 9 to 10:00 I had a daily ritual where I would write, usually after some meditation, so I could clear my mind from work. I'd finish up all my emails and Slack messages from work, do a little meditation, do my content, head to bed. That was my ritual. So I grew this — I think at some point I was the number 11 ranked person on LinkedIn, just based on the amount of impressions I'd had, how many likes per post I had. At that time, LinkedIn, it was all I was doing. But Paul said, you know what, you need to do this paid newsletter thing. So I did that, and having made this amazing connection — he, by the way, is the author of Product Compass, 70,000-plus newsletter subscribers, 165,000-plus on LinkedIn, he's huge, a great friend of mine, so you should have friends who are in similar positions as you — and he said, "Aakash, your writing is very suited for this," because at the time I used to write these five, six thousand word articles, drop a paywall at word 1,500 — the people who are only ever willing to read for five minutes were never going to pay for your newsletter anyways, but the people who want more, those are the type of people who are connoisseurs of good information, they're connoisseurs of research, they're willing to pay. So immediately found product-market fit with this paid subscription — it was wild, people were recommending it to their friends, I'm not even talking about it and it's just growing on its own. So I got this paid newsletter going, and that showed me, okay, I have this product, I have this side hustle that's working for me — in fact, the revenue I'm getting from this side hustle is going to be a lot more than I'm getting as a VP of product. So I was in that lucky position, unlike a lot of people who just have to leave their job and have no revenue — especially because I had kids, that was never going to be an option for me. I already had a good revenue plus line of sight to much more if I were able to spend more time.

Shobhit: I mean, I really admire folks who've built this career, their company, around deep writing — you, Lenny, Ben Thompson, who actually was an MBA classmate of mine, and I just saw his awesome writing, I really admire that, because I don't have that, or I haven't built that skill set of deep long-form writing that you have. It's really awesome. Now I'm also curious — as you're at this point thinking about it, or organically you come to that, but why writing? Did you consider coaching? How did you find out what was the thing that gave you most joy, and then just decide to pursue it?

Aakash: So I thought about the different options — if you think about how creators monetize, coaching, courses, books, YouTube, podcasts, writing — where I started was my bread and butter, what was I doing for free without getting paid, just for fun, that was that deep writing, so that's where I started. Everything more recently I've started to move into those other directions — released a book last year, this year I released a podcast. I'm kind of just taking the predictable creator path — a year or two from now there'll be merch, a year or two after that there will be an in-person conference. The roadmap is well-trodden, other influencers have basically already shown us the way, and we see how much revenue they have — you look at a Steven Bartlett, this man's blowing in like $50 million a year, so you see, oh wow, I could 100x my business, still there's a lot more room to go.

Shobhit: So I mean, I do think this is a very critical point, that both in careers and entrepreneurship, you don't have to reinvent everything, there is a roadmap, there's something other people have done and you can just focus on your — I don't know if you follow or use the term "zone of genius" — and just follow the other path for other people, at least I'm a believer in that.

Aakash: Yeah, I think that the haters on Reddit, they would say — like the PWS, the OG cases, there's basically two other paid newsletter editors, Vinamre and — I'm just blanking on the other name, Peter Yang — so like the four of us, they would say, oh you guys are copying Lenny. But no, actually — Lenny pioneered the category, yes, but it's not like in podcasting nobody else ever released a podcast, once podcast was created as a really good category a bunch of people entered, and actually Lenny co-created the category alongside Substack and GGO Rose and some of those other people, so he was one of the founding members, we're kind of all in the second wave of these new paid subscribed newsletters. It's fundamentally a space that has so much demand — if we think about how many PMs there are in the world, there's on the order of two to five million PMs in the world, Lenny definitely doesn't have 400 to 500,000 paid newsletter subscribers, otherwise he'd be living in Monaco in a high-rise, so he doesn't even have 10% of the market really yet. There's a huge ocean out there, and if you think about it, at least 10% of the market would be willing to pay for something like this. So we have a ton of space to play, and none of us have even started to see any sort of limit on the market demand.

Shobhit: I mean it's funny, there's a recent graduate of mine who was like, hey, one of the things I'm really thankful for is you told me to go subscribe — at that time it was Ben Thompson and Lenny — and this person is senior, so I totally agree with you about the lack of market penetration, because I believe everyone should be spending a fair amount on their education, this person was not doing that, and they're very experienced, really smart on that front, and it's a great investment for any product leader. Like, Lenny's newsletter, it's $150 a year, versus they're probably spending like $1,100 on a Reforge membership — go take a look at your login stats, how many times have your PMs logged into Reforge recently — or they're paying $1,000 for a fertility benefit per year, paying $600 to send people to conferences, what are people getting out of those conferences, I'm very curious to know outside of inspiration. So it's an interesting, very cheap thing —

Aakash: Oh, you mean for companies?

Shobhit: Yeah, yeah.

Aakash: I think most companies — my prediction is that within five to ten years most companies will have a learning and development budget. It was going to be like 100% of companies a year or two ago, I think the end of the ZIRP era kind of slowed the expansion of those, but those were on a crazy upward trajectory right before the bubble popped, and I'm sure it'll continue in a little bit.

Shobhit: Now I'm also curious, like, how do you decide what you're going to write about, and just also curious about your process — I know you often speak with other people, what do you go through to produce these awesome articles that people love?

Aakash: Yeah, so what to write about, and then my process for writing it, these are somewhat joined at the hip. So my fundamental thing I look at for how to write a post is what have people been asking me about the most, where I've been repeating myself the most. I have an engaged Slack community for paid newsletter subscribers, so they'll be asking me — I might look at like 10 or 11 questions a day — then on LinkedIn and X, people will comment questions and respond to my posts, a lot of people also DM me questions on LinkedIn. So all of these channels I'm getting questions from, and as I start to repeat myself, then I know it's time to do a newsletter. The other angle I use to write a newsletter is where is my curiosity taking me, and where is there no good content yet — one topic where my curiosity was taking me is what is good retention, it turns out Lenny has already written a really good post on what is good retention, so I haven't gone out and done the extensive research that would be needed to support such a piece, at least not yet — now his piece is getting three years old, I might actually do one maybe early next year, it's on the content calendar right now. But that's how I think about things. So for instance, I know we're going to talk about the product leadership job search — I didn't see any good content out there, there was a ton of content for early-career and mid-career PMs, but what about late-career PMs, and so it naturally made sense for me to dedicate — me and Colin, Colin dedicated like roughly two weeks to that post, I dedicated like a week to that post, and we collaborated to create the product leadership job search piece. Colin talked to something like 16 people, I talked to something like six people, and the idea is nothing I publish do I want to already be online, or that an AI could produce for you. These days AI is pretty good — it doesn't need to sound like an AI anymore, especially if you go try the latest model of Grok or the latest model of Claude Sonnet 3.5, nobody can tell the difference — in fact there was a recent Turing test done with Claude Sonnet 3.5, and the AI sounded more human than humans, so everybody was saying "this was written by AI, this was written by AI" — no, no, no, that was just written by below-average humans, and anything now that's kind of below average, we just think it was written by AI — no, no, it can fool you. So for me it's then, well, what are the unique insights the AI is not going out there and interviewing people, the AI is not going out there and doing research, and that's where I've really tripled down my content.

Shobhit: Totally love it, and I am also consistently amazed how much I can get AI to start sounding like me, by training it, and then be like — literally at the gym I would be thinking of something and I'd be like, ChatGPT, please come up with the draft, and I'm like, okay, it's 80% there, now I'll edit it later on and get it somewhere. That part is also super amazing.

Aakash: Yes, it can get you 80% there in like 10 seconds, and so your timeline to get to 100% is a lot faster.

Shobhit: I agree, I totally agree with you. Okay, that's awesome. So let's dive into two areas — in my community people have been posting your articles for a long time, but two of them really blew up and everybody was commenting on the things, so I wanted to talk to you about some of them. First, let's start with the product leadership job search — if you were to think of, as you interviewed these people, what were some of the most counterintuitive things you learned?

Aakash: Yeah, so that's generally my bar for publishing even ten words on my newsletter, and this is a 20,000-word article, so I'm going to choose the best of the best of that article — maybe the three most "wow, aha" moments for me. One, 70% of product leadership jobs are never listed online. Two, a similar percentage, about 70%, of product leadership jobs as a result are filled via recruiters, and so having a really good relationship with recruiters is very, very important. And then third, the salaries and comp percentages people are getting actually blew my mind — I think even when I was VP of product at Apollo, I thought I knew what CPOs were getting paid, but I was off, I was under, because there's a lot of data out there, but all that data is wrong, it's way, way under. The data is telling you these CPOs at Series C are making like $300k and getting like 0.5% of the company — no, it's actually way, way more, the range is anywhere from $425 to $600 in cash, and anywhere from half a percent at the very low end to like one and a half percent at the high end. People are making way more money than you think at the top. Those were three of my big ahas.

Shobhit: Love it, love it. I'm curious going deeper into especially the first point — when you say 70% of product leadership jobs are not posted, is this primarily a certain stage of the company, or is it across the board?

Aakash: Yeah, so there are a few public companies that have a policy that all jobs need to be posted online — even those jobs, 70% of them are filled well before they're posted, there's either an internal candidate or there was a recruiting person. If you look at a company like HubSpot — I got the opportunity to work with HubSpot's head of people at Apollo, we made a bunch of key hires on Apollo's product team — his approach, he built five, seven-year relationships with hotshots in product, and then when he'd go to a company like HubSpot or Apollo he would sell them slowly, and he would make the leader sell them. He'd tell me, "Aakash, this is one of the hotshots, he's a target for your team, he's led product growth at Miro, he's led product growth at Gong, he's going to be a perfect fit for Apollo, so you should go get dinner with him" — and I wouldn't even be talking about having a job yet, it would just be slowly cultivating into it. That's how these truly excellent companies are sourcing excellent high-level positions. The percentages decrease — if we're talking about head of product, CPO, SVP, even at like a Google, 0% of those jobs are listed, the only ones listed are the ones required by HR policy. If you go all the way down to director at a Series A company, then only 60% of the jobs are listed, so we see way more jobs listed. The smaller the company, the lower the title, the more likely it is to be listed.

Shobhit: Got it, got it, that's super interesting. And speaking of HubSpot, I actually interviewed with them for a VP role way back when, and then they decided we're not going to invest in this area yet. And how did it happen — I knew the CPO, the CPO was like, "oh, I read this article you wrote, really cool, want to chat about it?" I'm like, cool, that's awesome, let's do that — comes over for lunch, and it's like, oh by the way, we're hiring this person, I'm like, okay let's talk, it's a great company, I would have loved to work there. Now, the second point, building those relationships with the recruiters — there is a perception, at least I've sensed from my community and a bunch of others, that oh, recruiters are very transactional, unless you're a good fit then it's not going to work. I'm just curious what you found out in your research and writing on this topic.

Aakash: Yeah, I didn't find them to be transactional at all — just as an example, almost every single recruiter I reached out to as a part of this piece responded to my questions within that same day. I go out and do research, I send cold DMs every single day, and no class of people responded faster, more timely, than recruiters. So to think that they're transactional is a very broad mischaracterization, and if we're talking about the recruiters who are filling high product jobs, these people are incredibly well-paid — you could think these people are as good as your best CPOs and VPs of product. These are the guys you see at the yacht club on the phone talking about 800,000, this 0.1% of the comp — that's who they are, those people are called "million dollar men" because their goal is to make a million dollars in comp, and if they hit that by August, they'll just stop working for the next four months. That's cool, that's how these guys operate — a lot of them are gals, actually, a lot of them are women. No, definitely not transactional — all the best product leaders, everybody you know who's gotten a CPO job or SVP job has a network of recruiters, and to varying degrees they all support that network. The most common and effective tactic to use is that a recruiter is going to ping you — "hey, I think you'd be great VP of product at HubSpot, but you're currently not looking for VP of product at HubSpot" — so what you do is you text three of your buddies, "hey, are any of you guys interested in this VP product at HubSpot job, I have a lead" — one of them responds right away, you say, "okay, can you just customize your resume, I know they really care about this," she sends you her resume, you forward that to the recruiter, and you do this within like two hours, you don't wait 48 hours to do this extended process, you immediately come back and recommend her. And it turns out she gets hired for the job, because you actually know who the good talent is and you can recommend, and you can become that person who — when she got hired that recruiter got paid 80k, they got paid nicely, they're going to remember you after that, they're going to ask you again and again, and you can maintain these relationships. Then at some point you'll be the person they actually ping when they get new roles, and at some point that new role will be your role — you'll say, "you know who's the candidate? Me. I wanted to always work for Meta, and now you have a director role, and you know that's the level I'm targeting, and I've never made that seven-figure comp, here's my chance, let me talk with them."

Shobhit: If I were to boil this down, I think it's a very important insight — it's like, you've got to play the long game here, and you've got to just be willing to help and know it's going to come back to you.

Aakash: Yeah, yeah, I mean with karma, with these types of universe things, where you're just helping the universe, you have to put out ten times more than you're going to get back, you just have to be okay with that ratio.

Shobhit: Yeah, I mean, I don't know, in the long term it feels like for most people you get back even more, but on a short term —

Aakash: Yeah, like, don't expect it and tomorrow they'll have the job. And if you maintain a relationship with ten recruiters, you're only going to have like seven or eight jobs in your career, like three or four of them might get filled by those people, but more likely than not just one or two of them will. But that's great, that you got one or two of your key jobs filled by a recruiter — the other ones should come in just inbound, because you're developing such a great background. Anyone who has a good background in product knows that you're getting inbound — if you're not getting inbound in product, there's one of two problems: one, the brand, the companies you're working for aren't sexy enough, or two, you don't have enough years of experience. That's it. I've seen people with the least optimized LinkedIn ever, but they've been an Uber PM for seven years, they get inbound every week or so.

Shobhit: Agreed, totally agree. So let's talk a bit more about a recent article you wrote about AI in product management — again I'm going to ask you the same question, because you think the same way — what was surprising as you did your research and started to write things out?

Aakash: Yeah, so again this is another one of those, I think like a 9,000-word article, I collaborated with Marily Nika, who was an AI product lead at Meta and Google — she runs a Maven course as well, she basically knows the AI PM job market. I basically tried to get every insight out of her that you might get if you were to pay her for many, many coaching sessions, and then I also coached a bunch of PMs on getting AI PM jobs. This piece took me I think seven and a half months to create — you'll have to fact-check me on it, I say in the article exactly how many people we got PM jobs, but we got people jobs at Databricks, people jobs at OpenAI, Anthropic, Google, Meta, and of course the biggest hire, Microsoft — so many people getting Microsoft AI jobs, because we know Satya knows how to invest in a new area, and he laid off all those people in the gaming division to fund it. So there was the first "aha" moment for me after doing all that research — there's absolutely no concern about AI for product management, there's literally none. Everybody was worried for a second there because GPT-2 to 3 to 4 in two years was amazing, I agree with everybody — but you know what, GPT-5 hasn't come out, GPT-4o is not a big step up, code interpreter is not a big step up, we have not seen a GPT-5 level advancement — even though I did name-check Grok and Sonnet, which are two really competitive models now, even the latest version of Gemini — I teamed up with Liam Bolling over at Google, and even that's very competitive and very interesting, but none of them have been a step change. In fact, as we were just talking about earlier, what they're really good at is taking you from 20% to 80% — they can't do the beginning and they can't do the end, they kind of suck at both, they're not creative and they don't sound like humans. So what is AI useful for? Well, first of all, it's making your developers more efficient, second of all it's making your designers more efficient, third of all it's making you more efficient. PMs are using it for PRDs, for strategy documents, for responding to emails, for taking meeting notes, for sending out meeting summaries, for doing analysis on experiments, for creating bandit algorithms to better quickly select the winning A/B test scenario. So that's what PMs are using AI for today, and you know what it's meant — there's more PM jobs. There's this crazy data that came out, where in 2021 PM jobs peaked after 30 years of nonstop growth, PM jobs peaked, but you know why it's not down 10%? Because actually like probably 10% of PMs were laid off — it's those AI jobs that came in, and that's why it's only down 1%, because we saw basically like a 9% lift in AI jobs at these companies I mentioned, like Databricks, Microsoft — all these companies are hoovering up tons of talent. So that was my big aha. When I had that aha, my next question was, okay, all up, PMs who are laid off, or all up, PMs who don't have AI experience, how can you position yourself for one of these AI PM roles? So what I learned at the first aha is AI is the one place where side projects work — I don't think side projects work super well for your average PM job, if you've worked in retail and you want to move into fintech, you can't suddenly build a fintech side project. You know what, you could build a pretty good AI side project, or you could at least build something that shows you're really using the space, or you could become a content creator. A lot of people, what they do is, as a new model is released, they'll create a cool use of it, and then they'll get caught up in these Twitter and X threads where they compile the greatest uses of all the new models. You see lots of people like this — Omar Rizwan, who's now head of design at ElevenLabs, for every cool new model he used to release a little usage of it, so he went from being a regular designer to head of design at one of the hottest companies in the world. I saw many other people with very similar trajectories, where they're using their side project, they're hacking on the side, they're hacking with the AI, they're the AI hacker, they're using AI to build an AI product — those people are often getting these AI PM jobs. So that's route one. Route two — even if you're a core or a growth PM, how can you use AI in your surface area, how can you use AI in an innovative way — not just calling the OpenAI APIs, because there's absolutely nothing that's innovative or interesting about that. But maybe you fine-tuned the GPT-3.5 Turbo model to be more effective than GPT-4o and faster, so that you can deliver for your company at 1% of the cost in twice the speed, really good AI results for your users, that 90% of users accept those AI results — that's something that's very interesting. Believe me, to get to 90% user acceptance, that means you spent years iterating, improving your prompts, improving your fine-tuning, everything like that. So if you go and build those features, that's another really good route. Over the six, seven months that I was working on this piece, I actually worked with real PMs where we would try to build, like proactively think, in our roadmap, how are we going to get an AI feature, so that then when we go interview at Anthropic or OpenAI we can talk about how we are on the cutting edge of innovating product with AI.

Shobhit: What you just mentioned here, it seems for me to solve two big things that product managers always talk about — one big thing, and then there's a second point that's very important. First big thing is, when you're talking about "hey, I'm too busy," this is your opportunity to find that time for strategy, because you can leverage AI to do some of those things that you would have just spent hours doing, and those would be like the low-importance stuff for you. And then second, becoming a super user of AI seems to be just core for you to have even a ticket to entry to get an AI product management job.

Aakash: Oh yeah, and I think you have to code also — every single AI company out there, they want people who are technical, who can code, who can ship. I believe every single OpenAI PM that I know has gotten hired can code. If you want to work at probably the best tech job there is — you know that's probably going to be a two, three trillion dollar company at some point, it's only valued at 100 billion right now, so you're going to 2030, extra equity plus their cash offers are higher than Google and Meta — I mean if you want to work there, that's it, it's the sexiest job I believe currently in tech, you have to be able to code, I think it's a prerequisite. I tell everybody, Andrej Karpathy has a YouTube course, he helps you code GPT-3, everybody should be doing that.

Shobhit: Wow, that is really next level, and it feels like the product managers who are getting into AI have many levels to go through in their technical expertise, but they don't have to necessarily hit that top level right now to have the initial move, but like for something like OpenAI or Anthropic you need to get to that level.

Aakash: Yeah, ideally I think you should think about it as, how am I going to find two to three hours a day for the next six months to hack around with AI products, release AI products too — go ahead and take a look at levels.io, this man has, what, three AI products currently generating over $50k MRR. How many do you have? You know, everybody listening, generate even one with $500 MRR and you'll learn a lot — this man is releasing one a month, Marc Lou is releasing one of these once a month, there's no reason you can't also. The current state of Cursor plus Claude Sonnet is amazing, you can build apps fast, and as a PM it's really good for you to understand design as well, so use Playground AI to create your first version of your design, take it over to Figma, learn your Figma skills, improve it a little bit, and then — amazing.

Shobhit: Well Aakash, this has been such amazing information, I was taking notes while talking to you, I learned so much today. Thank you so much for appearing in today's podcast, and for our viewers, how do they learn more about you, how do they get to know some more of your content, tell us more.

Aakash: Sure, yeah, so I think the two ways — number one, subscribe to the newsletter, even better if you want to subscribe to the paid version, and if you need a discount of some sort, just email me, you can just reply to your welcome email, I read all my welcome email respondents, actually I read all my email respondents. And then second, listen to the podcast — just launched about a month ago, twice a week podcast, going like two and a half, three hours each recording session, going super deep on product topics. I was already doing this research already, and so all my friends kept telling me, if you're already doing it why don't you just put it out on the internet.

Shobhit: Love it, thank you so much Aakash, really appreciate it.

Aakash: Thanks for having me, and look forward to the next collab.

Shobhit: Hey, be sure to check out our website at intentionalproductmanager.com to see how you can level up in your career.