Hugging Face’s 12 free AI courses make more sense in this order
socialwithaayan · x · 2026-07-25
A practical order for Hugging Face's 12 free AI courses
The post argues that Hugging Face's twelve free courses are hard to navigate unless you know the right sequence. It recommends starting with the LLM Course, then moving to the Agents Course (with smolagents, LlamaIndex, and LangGraph), and then the Context Course for code-agent context engineering.
After that, learners branch into the area they actually want to build in: diffusion for images, audio for speech, computer vision, deep RL, robotics with LeRobot, post-training via the smol course, ML for Games, ML for 3D, or the Open-Source AI Cookbook when they want a hands-on notebook.
It also highlights that the certification is completely free. For the Agents course, you can audit it anonymously, earn a fundamentals certificate by finishing Unit 1, or get a completion certificate by doing Unit 1, a use-case assignment, and the final challenge. The Audio course uses a similar model. All hands-on work runs in pre-configured Spaces, with only basic Python, internet access, and a free account required. A bonus detail: Unit 3 of the Agents course includes building an agent that plays Pokémon battles.
More from coding & agent
- After four days stuck on GPU-mode Cholesky, the author retries with Opus 5 — dejavucoder · 2026-07-25
- Users report OpenAI Codex is down with 503 Service Unavailable errors — uwukko · 2026-07-25
- Claude Code command cheat sheet maps setup, review, and automation commands — Aiden_Tech_Ai · 2026-07-25
- Open-source repo bundles hundreds of AI attack and red-teaming tools — Aiden_Tech_Ai · 2026-07-25
- AI buzzwords in the next 30 to 60 days: agent graphs, headless AI, and more — arieljalali · 2026-07-25
- Modern Relay argues agent coordination needs typed graphs, not just Postgres — sarahdrinkwater · 2026-07-25