Increase your slope
A former Stripe new grad shares how to ramp up fast at the company: pick a narrow area and become its go-to person, then take ownership of small problems and build from there. It leans heavily on how AI tools should and shouldn't fit into learning.
- Become the team's mini-expert in one narrow area first; the author became the point person on Dashboard charts within about four months and kept that value for over a year and a half.
- Learn to ask better questions: use LLMs for "what" questions and people for "why," and come with what you've tried and a hunch before asking for help.
- Use AI to speed up execution, not to replace understanding; don't ship code you can't explain, and slow down in your first months to absorb how systems break.
- Take ownership early by fixing small frustrations yourself and prototyping fixes in an afternoon, then demoing them to stakeholders rather than waiting for approval.
Vikram Singh, who has been at Stripe for about a year and a half, distills his first-year advice and the lessons of other former new grads into three linked habits: become an expert in something small, take ownership of the problems around you, and get comfortable holding opinions. His core claim is that you can start all three far sooner than you'd expect. Expertise comes first, and it starts with whatever project you're handed. His own spin-up task was revamping the charts on the Stripe Dashboard homepage. He didn't find it fascinating, but charts turned out to be hard, and after three or four months he was the team's mini-expert. That status let him lead mitigation for chart-related incidents, work through dozens of run tickets, and weigh in on product and design calls, and a year and a half later his chart knowledge still matters on a different team.
The practical advice for building that expertise is concrete. Read beyond the architecture docs, including the product brief or UX research that justified the feature. Map out who knows which system and who remembers why decisions were made. When you ask for help, arrive with what you've tried, why it failed, and your hunch, and then ask the person how they reached their answer, since their intuition is what you're trying to absorb. Use LLMs for "what" questions and save people for "why." He also recommends dogfooding the product with test accounts and knowing the relevant metrics in Hubble. His example is useful: looking at Hubble data on the charts project revealed heavy "default dominance," with almost no users changing their settings, and that finding shaped feature priorities and still informs how he thinks about products.
On AI, his position is that it can speed up or hollow out your learning. Models make it tempting to let them do the thinking, which he calls a recipe for "long-term mediocrity." His advice is to slow down in your first month or so, use a coding agent inside an IDE so you can see and absorb each change, refuse to ship anything you can't explain and defend, and treat code review as a learning tool rather than a checkbox. Ownership, he argues, is ultimately about caring: understanding users, feeling responsible for what you ship, and weighing long-term consequences. You don't have to wait to earn it. Start with small problems you can fix alone, like a bug you spotted, a legacy testing library, a broken dashboard, or an obvious quality-of-life gap, and let each fix raise your own bar for quality.