Vladislav Kruglikov

LLM inference and infrastructure research engineer with 4+ years of experience in speculative decoding, KV cache efficiency, serving, and training-aware optimization. I have led inference, model training, and production infrastructure projects improving throughput, latency, and model delivery speed. I write a technical blog at https://vladislavkruglikov.com about efficient inference and training, speculative decoding, serving tradeoffs, and practical systems work. I have also served as a technical interviewer for algorithms, data structures, and NLP roles. You can find my CV at https://vladislavkruglikov.com/files/vladislav-kruglikov-cv.pdf