Yuri Pocepaev

Software engineer and lead developer at Neuroprem. Working on LLM systems: quantization, distillation, efficient inference, retrieval and memory for agents. Currently preparing a custom quantized GLiNER model for release and experimenting with Qwen inference and distillation. Technical notes, benchmarks and lessons from real systems. Engineering log: https://t.me/badbat4560 Articles: https://badbat4560.hashnode.dev DEV: https://dev.to/badbat4560 Code: https://github.com/badbat4560 Neuroprem: https://neuroprem.com Organization: https://huggingface.co/Neuroprem Less hype. More engineering. Instagram: https://www.instagram.com/badbat4560/