agents on Machine Learning Stuffshttps://moisesvw.github.io/tags/agents/Recent content in agents on Machine Learning StuffsHugo -- gohugo.ioen-usWed, 30 Sep 2026 12:00:00 -0500Loop Engineering: Design the Feedback Cycle, Not Just the Prompthttps://moisesvw.github.io/2026/september/loop-engineering/Wed, 30 Sep 2026 12:00:00 -0500https://moisesvw.github.io/2026/september/loop-engineering/Harness, loop, and graph are not the same layer. Design the middle one so agents improve a process instead of burning tokens on “try again.” The failure mode I keep seeing An agent gets most of the way there. The first draft is close. The tools work. Then it either declares success while the tests are still red, or retries the same broken step until the budget is gone.