Vladislav Kruglikov

I'm a research engineer and writer. I work at T-Bank on LLM inference and infrastructure. Previously, I worked at Yandex. I've been coding for 8 years and teaching occasionally.

I'm finishing my BS in Applied Mathematics and Informatics at HSE University, planning to continue with an MS and PhD. I also had the opportunity to co-author a paper accepted at EACL 2026.

Some of my favorite writing includes:

• KV Cache Pruning (Part 1)
• KV Cache Pruning (Part 2)
• Speculative Decoding from First Principles

You can read my writing or code, or follow me on LinkedIn. I'm always happy to talk about LLMs, inference, efficient training, and infrastructure. Reach out if interested.