# Tuning Matt Pocock skills for Machine Learning [Recently](2026-09-27%20Fine-tuned%20decoders%20for%20event%20extraction.md), I built an entire project and produced good results with Claude Code, running the Opus 5.5 model. This post describes my Claude setup for that kind of Machine Learning efforts. The biggest insight is how well [Matt Pocock's skills](https://github.com/mattpocock/skills) support machine learning work. I topically start all work with an interactive Q&A session using Matt's `grill-me` skill. For the last project, I then asked the model to write the [original DESIGN.md document](https://github.com/fnl/fine-tuning-decoder/blob/2f90282701bc573fedd93d24d7226a047506ec9c/docs/DESIGN.md). Next, to reach each milestone, I used the [`wayfinder` skill](Tech/Agents/skills/wayfinder/SKILL.md) to plan the route, and then the `implement` skill to drive the coding. Particularly the `wayfinder` skill stood out to me, and I tweaked it even further. I use [my own light variations of Matt's skills](https://github.com/fnl/agents), tuned to my needs. For example, I wove the [Karpathy Guidelines](https://github.com/multica-ai/andrej-karpathy-skills) into my `implement` skill variant. Generally, my versions are just a tiny bit more focused on testing, validation and scientific approaches. The AI assistant did virtually all of the work across the codebase and the browser, accessing the Colab notebooks and Weights & Biases score cards. The whole project took only about four workdays to develop and complete. And I maxed out the 5h usage window of my Claude Pro account on four occasions. That might be close to the weekly cap Claude accounts have. The project at some point ran into a classic snag: At first, the model generated nothing! Turns out, Unsloth's copy of the chat template for Qwen3-4B adds an empty thinking block that the model cannot handle. Encouragingly, Opus was pretty quick to identify the culprit on its own. Overall, Opus 5.5 needed only very occasional directional input from my side: Most of its choices and decisions were reasonable. The project is another showcase of the huge progress autonomous copilots have made since the start of 2026.