AI Automation
Practical AI automation for products and operations
I design AI assistants, internal tools and workflows that remove real manual work - in support, operations, research and analytics
Problems I work with
Where AI earns its place in the stack
Support teams overloaded with repetitive requests
Manual operational workflows that don't scale
Analyst time spent on data prep instead of analysis
Internal tools built on spreadsheets and tribal knowledge
Unclear where AI actually helps vs. adds risk
Prototypes that never reach production
How I think about AI
Practical AI implementation, not abstract AI enthusiasm
Workflow first, model second
The product around the model matters more than the model. Most AI wins come from rethinking the workflow.
Shippable, not impressive
I focus on AI features that hold up in production, integrate with existing systems, and people actually use
Operational realism
Error handling, escalation paths, observability and cost - designed in from the start, not bolted on later