Bartosz Mikulski is an AI and data engineer whose specialty is the gap between a working demo and a dependable production system. He moves AI projects from the good-enough-for-a-demo phase to production by building testing infrastructure and then fixing the issues those tests detect, an unglamorous discipline that is the difference between an impressive prototype and software people can trust.

His engineering toolkit reflects that focus. Bartosz designs robust data pipelines and applies testing strategies such as snapshot and integration testing, working with data quality tools like Great Expectations and Soda alongside SQL and Spark tests. Beyond testing, he works on the full productionization problem: pipeline orchestration, prompt optimization for stability, and caching strategies that reduce load and improve responsiveness, so that models behave predictably once real users depend on them.

Bartosz’s career path took him from Java development into data engineering and then into AI engineering, and he documents what he learns along the way through near-weekly writing on his blog. He shares his knowledge generously: he teaches programmers and non-programmers alike how to use AI, contributed a chapter to the book 97 Things Every Programmer Should Know, and has been a speaker at several conferences, including Data Natives, Berlin Buzzwords, and Global AI Developer Days.