I landed a job at an AI startup. Here are my tips for working in AI.
Summary
Lambert Liu, 22, a software engineer at Graphite (an AI code-review platform), explains how he chose a startup path after interning at Google and Replit. He contrasts the stability and training you get from Big Tech with the fast learning, ownership and impact available at startups. Liu shares practical advice for graduates and early-career engineers hoping to join AI startups: leverage Big Tech internships, build projects, practise interview fundamentals, learn system design and become a holistic engineer who takes ownership.
Key Points
- Big Tech internships demonstrate a strong technical foundation and reliable delivery — they’re valued by startups.
- Startup experience helps, but you can substitute it with well-executed personal projects that show autonomy and problem-solving.
- LeetCode-style algorithm practice still matters for interviews, but product thinking and ambiguity-handling are increasingly important.
- System-design thinking can be asked even of new graduates; courses like human–computer interaction can help frame problem scoping and trade-offs.
- Be a holistic engineer: move fast, take ownership, care about users and produce high-quality work under ambiguity.
Content summary
Liu describes his path: two internships at Google followed by a stint at Replit, which convinced him startups offered steeper learning. He argues Big Tech experience signals solid engineering fundamentals to recruiters, while startup roles reward adaptability and ownership. He emphasises that personal projects—AI or otherwise—are strong signals of initiative and problem-solving. Interview prep should include algorithms plus practice with ambiguous, product-focused questions. He also highlights system-design thinking and recommends courses and hands-on projects to build that perspective. Above all, startups want engineers who move quickly, own work and care about users.
Context and relevance
This piece is useful for new graduates and engineers pivoting into AI roles. It reflects hiring trends where technical competence (often proven via Big Tech internships or projects) combines with product sense and the ability to work in ambiguous, fast-moving teams. For anyone weighing Big Tech versus startup paths, Liu’s experience illustrates trade-offs: structure and polish versus ownership and rapid learning. The guidance aligns with current industry demands for pragmatic engineers who can ship and iterate quickly while understanding product impact.
Why should I read this?
Want a fast, no-nonsense checklist from someone who actually did it? Read this. It’s short, practical and full of sensible tips you can act on right away — build projects, practise system design, and show you care about users. We saved you the time of wading through the noise.