How many years of experience should I rack up at Amazon / in Big Tech before moving to a startup?

In a vacuum, my recommendation is to stay at Amazon for at least 1.5+ years so no one questions your tenure at the company (I actually made a video about this topic here).

It's probably in your interest anyway to stay at the company for 2-3 years due to your stock vesting schedule at Amazon. So stay unless you really hate your team/manager, and even then, I'd first try to switch teams before exploring externally.

Once you have a decent stint at Amazon on your resume, you now have the "FAANG glow" which will open up more opportunities. So the number of companies that reach out to you will only increase, which means your chance of finding a good startup will be good.

Moving to a startup typically only makes sense if:

1. The product already has meaningful traction.
2. The team is very talented, which effectively means the worst case outcome is that you learn a lot and then get acqui-hired into a larger company.

Over-simplified answer: After 2 - 6 years

Less simplified answer: After you get to SDE 2

Longer answer: After you get tired of working at Amazon, have done some soul searching, and are certain you want to work for a smaller, more nimble company

Engineers who came from Big Tech (but weren't there for too long) tend to do well in startups. Every startup wants to become a Big Tech company some day, so coming from that background lets you teach others how to build software more responsibly, scale systems and culture, and just do things the right way.

Here's my advice for engineers going from Big Tech -> startups:

• Be comfortable being scrappy - At a company like Amazon, you're going to be used to A/B testing + logging everything, doing waves upon waves of technical reviews, and just taking a lot of time in general to ship everything very carefully. You cannot bring this mentality to a startup (this is why engineers who have been at FAANG forever often do poorly at startups). Startups have to move fast or they die. Of course, you should still care about code and product quality, but you can generally only get the low-hanging fruit there with early startups. Don't even bother doing things like setting up extensive logging and automated testing suites - Dogfood the product yourself and directly talk to users.
• Don't wait for others to tell you what to do - This is because startups themselves don't entirely know what they're doing: They're trying to figure it out! Unlike Big Tech, startups have literally infinite scope. Engineers in startups are rewarded far more for taking initiative, being creative, and creating their own projects. Take advantage of that.
• Get to know everyone - It is impossible to feel connected to everyone in a company like Amazon: It's simply too big. One of the unfair advantages of an early-stage startup is that everyone can actually get to know everyone. Take the time to build up relationships, get to know your coworkers, pair program, and help others aggressive. Startups can build up organizational synergy that FAANG companies can only dream of. Check out our relationship building masterclass to help: [Masterclass] How To Build Deep Relationships Quickly In Tech

I recommend checking out these other resources too:

• "What's the difference between Big Tech and startups in terms of learning and growth?"
• "How Effectively Do Big Tech Engineers Perform In Startups?"