ML/AI Engineer (LLM, RAG, Agentic Tools)

Що очікуємо

• Have hands-on commercial experience working with vLLM in self-hosted environments, including model optimization on H100/H200 GPUs, with a solid understanding of GPU architecture principles and cache behavior under production workloads.
• Have a deep understanding of the RAG concept and a proven track record of successfully deploying similar solutions in production.
• Have worked with Haystack, LlamaIndex, LangChain, or similar frameworks, and understand the core differences between them.
• Have experience using MCP and working with LLMs that call external tools and services.
• Have strong Python skills, including architectural principles, design patterns, multithreading, asynchronous programming, modern testing practices, and experience operating systems under load (monitoring, memory optimization, telemetry).