Real Interview Experiences
Learn what to expect directly from candidates and interviewers who've been through it.
Interviewed recently? Share your experience to try it free, or become a member for full access.
“Good prompts from the interviewer, they want to move through metrics quickly so they can get to second prompt around tradeoffs ”
“So far straightforward phone screen. Prepped background and interests well and tried to speak to the role in a bit of detail. ”
“It was very clear overall and well supported by the recruiters. The only confusion was when giving the final presentation, I was told it was geared towards internal teammates. Then the interviewer asked me to present as if to executive customers.”
“It was very conversational. The recruiter call was pretty easy and direct. The next round was with two interviewers and I was given 45 minutes to dive deep into a technical challenge or program I solved or drove. It was very very conversational and while they probed it was not very adversarial”
“The problem was challenging. There were many conditions you had to satisfy and some aspects were a bit underspecified. The interviewers were stone faced and gave zero commentary. I ended up making a small bug which I wasted about 15 minutes trying to fix.”
“It was very rigorous but the recruiter was actually accurate and helpful about what to prepare, which is rare and I really appreciated that. The process consisted of a system design interview and a coding interview. The system design round focused on designing a payment processing system. The interviewer pushed beyond a basic design and spent a lot of time exploring failure modes, retries, idempotency, operational concerns, and correctness under real-world conditions. It felt closer to designing something that could actually run in production than many system design interviews I've done elsewhere. The coding round was also fairly challenging. The problem itself wasn't especially difficult from an algorithmic perspective, but it required careful handling of edge cases and a precise understanding of the requirements. I felt like there was less emphasis on LeetCode-style tricks and more emphasis on building a correct, production-quality solution. What went well was that the interviewers were engaged and clearly knowledgeable. The conversations felt collaborative rather than adversarial. What didn't go well, at least for me, was that I underestimated how much depth they expected around correctness and edge cases. In hindsight, I would have spent more time validating assumptions and thinking through failure scenarios before diving into implementation.”
“The process was short-lived for me, unfortunately. The coding interview was an hour long, and the question was similar to the game of life, but for an infection spreading across a 2D array. I approached it much too methodically - explaining my approach before implementing, like I've learned to do from past technical interviews. The thing about this one is that theres FIVE long parts, and you have to pass the test cases for each part before you move on to the next. If I could do it again I would be frantically coding from the very beginning - waste no time. Add print statements to debug. The interviewer doesn't help or guide much, its mostly on you.”
“The bar is pretty high, but the process is very smooth. The coding and system design need to be completed fast.”
“The recruiter was kind and well organized. The product sense question was unique and fun to solve. The execution question threw me through a loop. I assumed OpenAI carefully released models for safety reasons but was told they give the model to everyone immediately to test it, so I had to pivot my launch strategy at the last moment.”
Preview full access of other recent interview experiences
“The weirdest Anthropic round was the company values interview. It was almost like a therapy session, and honestly if you went to a therapist at some point, you will pass that round much more easily.”
“What was very unusual is they didn’t give me any tooling to draw the system design, so I just sketched it on a piece of paper and talked them through it, then we got into this oddly deep debate about whether hover-over history should count as a recommendation signal.”
