From big tech to startups — how product management differs based on size in the AI space
PM work differs based on both size and industry; we all know that. But what about AI-led product management specifically?
Product management is going to — if it isn’t already — be one of the most varied and ever-changing disciplines in tech. That we know. But to experience this all firsthand, I ventured off into both big tech, a medium-sized tech firm and a Series B startup. In this article, I’m going to dig into the differences in leveraging AI for product development and building an AI product in all tech company sizes.
AI itself is evolving rapidly daily— product management and its various processes are ever-changing. AI can now help with writing one-pagers, requirements, and automating everyday tasks. Alongside this are AI products themselves — PMing an AI product is different than a non-AI product. AI-native products focus on discovering AI use cases, customer conversations, security, trust, and rapid experimentation. PMs are now expected to vibe code with tools like Claude Code, Cursor, Loveable, and Replit. AI note-taking as well as AI-generated PRDs have become the norm. With all that said, let’s explore the pros and cons of working in each sector (big, medium, and small) based on what I’ve gathered from various interviews and experiences.
Big companies can be agile too, depending on team
Traditionally, outside of FAANG, folks would assume that large tech companies adopt innovative technologies for the sake of the board and to keep competitive, but are slower in execution. With a slower pace and flexibility as a PMN comes more stability; we all know that’s the main ‘big tech’ trade-off. This holds for most large tech organizations. Even FAANG companies, which sit on top of a large pool of cash to invest in AI, can move more slowly to ship and adopt. When I started my PM career at Microsoft, we had access to GPT3 back in 2022, pretty early on, before the rest of the industry caught up, but execution couldn’t match the pace of ideation. The only key point, though, is that it’s all team-dependent. If you have a fast-moving team in a large company, it can feel amazing: you move faster than the rest of the company with a safety net if things go south. Let’s summarize the benefits here:
PMs get less risk and more stability, and could potentially still move fast and work with AI well, depending on the team.
Potentially has a large pool of cash for AI investment.
Better salaries and benefits, as well as stable equity.
And in my opinion, the cons are:
Despite wanting to remain innovative, big tech tends to move and execute more slowly due to size and red tape.
The chances of gaining 0 to 1 product experience, especially if it’s an AI-native product, are slimmer.
Similar to the above: while possible, experiencing product growth at a massive scale is unlikely.
If you’re exploring the idea of working as a PM in a large tech company, there are obviously many aspects at play outside of their usage of AI:
Is their culture what you want and align with?
Are you passionate about the product and/or the space?
Are they offering you sufficient compensation and benefits?
A big company with a culture that enables fast AI exploration is still possible. You look at teams like Meta and Google, which prioritize AI-enabled services and teams while also using it to replace internal processes and tasks where possible. But in general, I wouldn’t be surprised if most large tech companies are still leaning on traditional processes without changing too much of their daily operations (why change something if it has worked in the past?). AI adoption and culture are likely team-dependent. Some teams may be vibe-coding MVPs for better discovery work to show clients or customers their product ‘vision,’ while others may be moving more slowly without using AI that much at all.
Mid-sized companies are a wild card.
Planview, a company with between one and two thousand employees, a couple of hundred million in revenue, with many products and a mature platform, is the epitome of medium tech. You have a company that’s experiencing some growing pains, but is still trying to be AI-forward. Of course, medium tech is a broader, subjective term. Some could refer to a company that comprises fewer than 50 employees as a medium-sized company if they already bring in over 50M in revenue and have been around for decades. Because of the rise in early-stage startups and the enablement of fast-growing ones due to AI adoption, the definition of medium tech could also change in the future.
In a medium-sized tech firm, I experienced both the pain of working with old legacy software that suffers from slowness, yet can’t be deprecated due to customer use, imbalanced team paces, and AI innovation. I believe you’ll see the same pattern across medium-sized companies: an interesting dynamic between PMs who drive the steering wheel like they’re in a startup and want to move fast and innovate, and teams who prioritize the current ship or move more slowly due to traditional company culture that’s been cultivated in prior years. This causes an imbalance that needs to be navigated carefully. That’s why I believe choosing a medium tech firm can be a wild card: if the organization is in a transition to adopt AI more rapidly while shifting its culture to reflect that, it can be a good sign. Similarly, that said transition can also be poorly executed, or the company itself is happy with its stable revenue and doesn’t move as fast to adopt new technologies. As a PM, it could be less exciting to move fast and experiment with AI, although I’m sure you’ll have other interesting work to do.
In my opinion, the pros of working in a medium-sized tech company are:
Similar to big tech, the risk and stability levels are comparable. You may also get more visibility in your work than in a big tech firm.
While compensation may not be as high compared to VC-funded startups or large tech companies, they’re still good and well-respected.
AI adoption is a wild card, but the team could execute faster than a big tech firm.
And the cons can include:
Imbalance within the culture, especially if a medium-sized company is older and comes from traditional backgrounds.
Medium-sized companies will prioritize growth and retention, which means there could be less innovation on the AI front.
Some medium-sized teams will definitely have adopted AI well, both internally and within their customer-facing products. Vibe-coding and generating one-pagers with LLMs could already be at play. But you never know — and that’s a key answer to seek while interviewing for new PM roles, no matter the size.
Startups offer faster experimentation with AI.
AI-native (or adjacent) startups move fast, are deep in discovery, and expect all their employees to tinker or build with AI. They live and breathe it. Even non-AI startups are exploring their various applications internally or pivoting to include its use in their existing platforms.
PMs in AI-based startups will expect to move faster, if not just as fast as non-AI startups, but the biggest differentiator is product discovery work. Understanding the barriers of building with AI will be tricky, as B2B startups will seek niche use cases, jump over security hurdles, and pivot fast (and vibe-code) within scenes of ambiguity. That’s reality.
Generally, startup land is thrilling; you have the chaos of experimentation and shipping MVPs fast; you have the environment for highly rewarding discovery work. Finally, you get to scale something from a baby to a massive success and accelerate your career ladder if things fall into place. Let’s summarize these benefits:
A good startup with strong funding or available cash (from revenue or elsewhere) can definitely make for a fulfilling PM career accelerator. Being part of a positive rocketship can feel beyond amazing.
AI startups are all about becoming AI-native, or shipping AI features fast with the latest capabilities. There will be nothing short of opportunities for you to explore and tinker with.
Autonomy and visibility are no longer privileges; they’re responsibilities. It’s a double-edged sword: when the business is winning, you’re the hero.
If you’re in or interested in product management and are still early in your career, I always recommend startup experience — the grind and the growth can be fun and fulfilling… when things are going well.
When things aren’t going well:
Being part of a sinking ship can feel worse than being laid off — I’ve had countless folks tell me about their startup struggles.
Being laid off isn’t uncommon in the startup scene, as everyone knows. They’re always looking for the best of the best, and anyone who isn’t aligned with their energy or performance might get the can.
PMs who get a lot of autonomy in a startup will be the ones responsible if the business collapses. I mentioned above that winning feels like you’re the main hero. Losing…feels like hell.
But lastly…
Join somewhere that is AI and tech-forward.
No matter what, there’s value in both learning, career growth, and connections across all tech company sectors and sizes, as long as it’s a tech-first company. Tech-first companies operate differently from non-tech-forward companies — the former are usually more innovative, have interesting tech to work with, are AI-friendly, and are caught up to speed with the latest ideas or processes.
While I have done internships and pro-bono gigs with startups before, I’m joining one full-time now that I have over 5 years of PM experience under my belt to fully witness the chaotic life of clearing ambiguity and achieving early growth. Joining any startup that has a limited runway can be daunting, yet exciting.
I went from Microsoft (100k+ employees) to Planview (1k+ people) to Unbounce (180+ folks), so the natural course of my career was to go even smaller and work to grow a startup to a billion-dollar valuation.
Whether that becomes a massive success or not will depend on the team, including me. There’s risk. But there’s also thrill and energy.
About Me
My name is Kasey Fu. I’m the co-founder of PM Hive and work full-time as a PM. I host the PM Hive Podcast, write for various publications, and keep a newsletter with over 3000 subscribers. Consider subscribing to the PM Hive newsletter today!