Bartosz Mielczarek

Member of Technical Staff, Engineering

Bartek is an AI Platform Engineer working across MLOps, infrastructure, and full-stack development. He builds the systems behind post-training and reinforcement learning pipelines, runs GPU compute clusters, and ships LLM inference and serving infrastructure. He also develops internal platform tools that make training, evaluation, and serving faster for the teams using them. His focus is on reliable, observable systems that make it easy to iterate on models from experiment to production. He is a coauthor of two Patronus AI research papers, one on an open source model that detects hallucinations in retrieval-augmented generation and one on a small evaluator model that grades LLM outputs against user-defined criteria. He studied mathematics at University of Warsaw.