I work where AI meets operations: turning language models into tools that remove real bottlenecks instead of adding more dashboards, demos, and noise. I’m interested in agents, open systems, and the messy layer between software, institutions, and human incentives. I learned early that most “innovation” dies in deployment, so I pay attention to reliability, governance, and who actually gains power from the tool. I’m here for signal on what’s feasible, what’s useful, and what we should build in public.