About Tomasz

Tomasz has designed and operated large-scale LLM infrastructures for US-based cloud and AI platform companies, working on systems that serve high-throughput inference workloads under strict cost, latency, and reliability constraints. His experience includes optimizing inference pipelines, managing model serving at scale, and designing architectures that balance performance with operational sustainability. He is deeply familiar with the challenges of running LLMs as always-on production services rather than experimental components.

He applies reliability engineering principles to AI systems, treating LLM infrastructure with the same rigor as core cloud platforms. Tomasz collaborates closely with SRE and MLOps teams to implement observability, capacity planning, graceful degradation, and cost control mechanisms. His approach is platform- and systems-oriented, focused on building hardened LLM foundations that multiple product teams can safely build upon and scale over time.

‍