Self-driving cars are still fun to work on. In the perception team, you can almost touch everything, from modeling, onboard integration, and model optimization to the data pipeline. So, you can learn a lot.
I still believe self-driving is one of the coolest things to apply ML to.
Management, not clear future roadmap. Uncertainty for scaling. Several teams are doing similar things.
1 phone interview and 4 rounds of onsite interviews. Asked general algorithm and design questions. Onsite interviews: 2 rounds in the morning, followed by lunch with the manager, then another 2 rounds in the afternoon.
Short introduction to yourself and interviewer, then start coding. Ask some questions to the interviewer at the end. Two coding questions in a round; be well prepared for edge cases because they really focus on it.
I applied online. The process took 2 weeks. Some interviewers are good while some are not. They expected you to state pros vs. cons on your solution. Some interviewers are really quite during the interview. You have to push the interview process on
1 phone interview and 4 rounds of onsite interviews. Asked general algorithm and design questions. Onsite interviews: 2 rounds in the morning, followed by lunch with the manager, then another 2 rounds in the afternoon.
Short introduction to yourself and interviewer, then start coding. Ask some questions to the interviewer at the end. Two coding questions in a round; be well prepared for edge cases because they really focus on it.
I applied online. The process took 2 weeks. Some interviewers are good while some are not. They expected you to state pros vs. cons on your solution. Some interviewers are really quite during the interview. You have to push the interview process on