Reflexion: Make an AI Agent Learn From Its Own Mistakes

An agent that fails a task and just… tries the same thing again is useless. Reflexion gives it a memory of its own mistakes: fail → write a lesson → retry with that lesson in context. It's learning without touching the weights. 🪞 Watch it fail, reflect, then nail it: https://dev48v.infy.uk/prompt/day20-reflexion.html The loop Attempt the task. Evaluate — tests, a checker, or a self-critique say pass/fail (with a reason). On failure, reflect — the model writes a short verbal lesson: "I assumed…