What You'll Do
- Deeply understand the highest leverage problem: Sit with agents and watch them work. Understand their economics, the workflows that waste their time, and where they reach for a spreadsheet or the phone instead of the client's product. Sit with product to turn that into a point of view on what to build next.
- Design and build it: Ship complete features across the stack: interface, API, data model. In some cases, you might define and ship an interface and partner with another founding engineer responsible for AI agents.
- Ship and learn: Get work into production fast, watch how agents actually use it, and iterate to help them improve their outcomes.
- Set the bar: Establish the shared components, interface patterns, and visual language that future features build on.
- Work AI-first: Use AI throughout your work: research, prototyping, design exploration, code, and testing. Help the client figure out where AI belongs in the product itself, not just in how the client builds it.
What Success Looks Like
Six months in, you've shipped dozens of features that agents use every day and would notice if they disappeared. You can point to specific workflows that are measurably faster or more effective because of your work. And the product feels more coherent than when you arrived, because the patterns you set are the ones the rest of the team builds on.