Notion's interview summary with Lovable cofounder Fabian Hedin presents using your own product as a way to stay groundingTying a model's answers to checkable sources — retrieved documents, live data, tool results — instead of letting it answer from memory alone.Full definition → in its actual capabilities. Lovable, which lets people build software through natural-language requests, used its platform for internal tools such as office CRMs and operations software. Those everyday uses kept the team close to what the product could do. The summary connects this experience with Hedin's advice to build around models' existing capabilities, after initially pursuing complex AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → systems that anticipated future improvements.
Terms in this piece · Glossary
grounding — Tying a model's answers to checkable sources — retrieved documents, live data, tool results — instead of letting it answer from memory alone.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
