Startup Spotlight: Proper Motion wants agents to learn from production
The founders pair OpenAI agent work with finance and infrastructure experience; Proper Motion says its private beta reached $270,000 in annualized revenue in three weeks.
By RuntimeWire Staff · Published
Primary source: a16z SPEEDRUN
Why it matters
If agents need ongoing corrections to stay useful in production, the layer that connects observed behavior to tested changes could become as important as the tools that build and monitor them. Proper Motion's early revenue and pilots are company-reported; measurable customer improvements would be the stronger test.
Proper Motion, a San Francisco startup co-founded by former OpenAI employee Rafit Jamil, is building software to help AI agents improve from real-world use. In its a16z Speedrun profile, Proper Motion describes a system that connects agent behavior to business outcomes, then helps teams identify and deploy corrections.…
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.