AI systems and reliability
AI verification, system evaluation, self-testing, and AI for data management.
Research agenda →Technical Expertise & Independent Analysis
Evidence-backed technical expertise in reliable AI systems, AI for data management, query optimization, cloud data infrastructure, and systems performance.
My work spans reliable AI systems, AI for data management, query optimization, cloud data infrastructure, and systems performance.
Across research and industry, I focus on understanding how complex systems behave, why they fail or become inefficient, and how their performance, reliability, and cost can be evaluated rigorously.
The foundation is published research, working systems, open-source software, teaching, and documented technical impact.
Evidence by technical area
SQL and query optimization
Query rewriting, equivalence, program synthesis, physical design, and execution behavior.
Research evidence →Cloud warehouses and FinOps
Workload-aware tuning, cost/performance tradeoffs, warehouse sizing, and learned system control.
Research evidence →Contact
For research collaboration or a technically grounded discussion, use the ordinary academic contact channel.
mozafari@umich.edu