8 min read How Gimlet Labs likely interviews compiler, distributed-systems, and GPU-architecture engineers for its multi-silicon inference cloud, and how to prepare. Read article
7 min read How speculative decoding speeds up token generation without changing model output, and the questions inference interviews use to test whether you understand it. Read article
7 min read What GPU and CUDA interviews at Nvidia, Groq, and the inference startups actually test, from warps and coalescing to occupancy and kernel fusion. Read article
6 min read Prefill and decode, the KV cache budget, continuous batching, and autoscaling GPUs, walked through the way the interview actually goes. Read article
7 min read A walk through Together AI’s engineering loop, from the coding rounds to GPU-serving system design and the CUDA-flavored take-homes. Read article
What's missing, wrong, or worth adding? One line helps.
Sent — thank you. We read every one.
