Did the OpenAI onsite (virtual) in January. Five sessions across two days. Here's the unfiltered version.
Structure:
Day 1: Coding round 1 (45 min, one problem, live in CoderPad) System design (60 min, one design problem, Excalidraw) Cross-functional / PM-adjacent round (45 min)
Day 2: Coding round 2 (45 min) Behavioral / values (45 min)
Breaks between sessions were short. By session 4 I was mentally running on fumes. This is a real consideration: the energy management across a 2-day virtual onsite is different from a 1-day in-person loop.
Coding rounds: Both were 'medium that escalates.' First one I got a tree problem that started as a traversal and ended with me implementing a custom comparator for a priority queue. Second was graph-based shortest path with a dynamic edge-weight constraint they added 25 minutes in. Both times the interviewer told me upfront: 'we're going to add complexity, that's expected, we want to see how you adapt.'
System design: I wrote about this separately but briefly: design a distributed rate limiter. Deep on the storage layer (Redis vs. a custom token bucket, why, tradeoffs), deep on how you handle edge cases at the API gateway layer. 60 minutes goes fast.
Cross-functional round: A PM asked me about shipping a feature with incomplete requirements. 'Here's a hypothetical: your team ships an experimental feature but usage data is ambiguous. How do you decide whether to invest further or cut scope?' This is not a coding problem. They wanted engineering judgment + product thinking. No right answer, but they noticed if your framework was incoherent.
Behavioral round: About 4 questions, STAR format, heavy on ambiguity and judgment calls. Not 'tell me about a conflict' generic stuff. See my other posts for specifics.
Debrief timing: Got a call from the recruiter 8 days later. She said debrief went well and asked about offer preferences (equity vs. cash balance). Verbal offer in the same call. Written offer 4 business days after.
Total loop to offer: about 7 weeks. Felt long while I was in it, reasonable in retrospect.
6 replies
staff_steph (Primly starter)
The 2-day virtual structure with short breaks is brutal. One thing I tell people: block time the evening between day 1 and day 2 to do nothing. Don't review notes obsessively, don't prep more. Just recover. Your brain is doing work in the background.
hardware_hugo (Primly starter)
8 days for debrief-to-verbal is fast by FAANG standards. Google and Meta both took 2-3 weeks on mine. Did you have a competing offer that accelerated anything, or was that just their normal pace?
de_derek (Primly starter)
No competing offer at the time. That's their baseline pace based on what others have told me. They seem to move faster than the traditional big-tech loops.
alex_design (Primly starter)
Did the interviewers tell you how you did at the end of each session or was it a total black box until the recruiter call?
frontend_fran (Primly starter)
The 'we're going to add complexity, that's expected' heads-up is thoughtful. Hate when interviewers add constraints without warning and you're not sure if that's punishment for being too fast or just how the problem works.
Primly Team
One stage people tend to underestimate in loops like this is the deep-dive on past work. In most senior-ish processes, it is less “tell me about a project” and more a structured technical interrogation: what was the objective, what constraints were real (latency, cost, privacy, safety), what you personally owned, and what you would change with today’s information.
A useful way to prep is to pick 2 projects and write a one-page “artifact” for each: architecture sketch, key tradeoffs you made, 2 hard incidents (debugging or rollback), and 3 metrics that actually moved. Then rehearse answering follow-ups in a tight loop: “What failed?”, “How did you know?”, “What did you try first and why?”, “What did you de-risk early?”, and “What would you not do again?”
For OpenAI roles specifically, be ready to connect choices to safety or responsible deployment, not just performance.
What’s the toughest deep-dive follow-up you’ve gotten, and how did you handle it in the moment?