Taro Logo

AI Software Engineer Interview Experience - Shanghai, China

March 1, 2025
Positive ExperienceNo Offer

Process

My objective is the AI algorithm engineer position at Baidu MEG. The whole interview process was quite challenging. It consisted of three main parts, each focusing on different technical skills and knowledge.

First up was the coding round, some typical LeetCode Medium problems like LRU Cache, Reorganize String, Sort List, etc. I nailed it, AK and bug-free.

Next, I met with another interviewer who focused on deep learning frameworks. We had a lot of discussion about PyTorch and TensorFlow, as well as my past AI projects. The interviewer was really nice, and we had an in-depth conversation about model training and optimization. It was a great exchange of ideas.

In the afternoon, the third interviewer grilled me on GPU architecture and optimization. This part was really tough - he asked a lot of detailed questions about hardware, cache optimization, and even PTX and SASS. Most of my work has been on the CUDA framework, so I didn’t have much experience with low-level GPU implementation. As a result, I didn’t perform as well in this section.

Overall, the interview process was smooth but highly challenging. While I didn’t pass, I still found the experience very rewarding.

Questions

First round is LC medium questions, then AI frameworks and project experience, finally GPU hardware details.

Was this helpful?

Interview Statistics

The following metrics were computed from 1 interview experience for the Baidu AI Software Engineer role in Shanghai, China.

Success Rate

0%
Pass Rate

Baidu's interview process for their AI Software Engineer roles in Shanghai, China is extremely selective, failing the vast majority of engineers.

Experience Rating

Positive100%
Neutral0%
Negative0%

Candidates reported having very good feelings for Baidu's AI Software Engineer interview process in Shanghai, China.

Baidu Work Experiences