Learning Path
Neural Mastery's 200+ pages aren't a flat list — they follow a real dependency order, across 7 sections and 28 sub-topics. This page has two views of the same structure: a dependency graph, and a full curriculum breakdown.
Have a specific goal already? Suggested Paths trims this same structure down to a shorter, role-specific route (ML Engineer, LLM/GenAI Engineer, AI Research Engineer, AI Systems/Infrastructure Engineer).
The Dependency Graph
- Each bar = your progress. Every page has a "Mark as understood" button at the bottom; this reflects that live, stored only in your browser — no account, nothing sent anywhere. (This page's own button is real too — it counts toward your overall total on Progress — it just doesn't move any bar above, since this map itself isn't one of the 7 sections it's mapping.)
- Click a node to jump straight to that section.
- Follow the arrows if you're not sure what to read next — it's one straight path, not a set of parallel tracks: Foundations → Models (by far the largest section) → Agents & Applications → Systems & Infrastructure → Safety & Evaluation → Research & Build → Career.
- If you already know a section (say, you're already strong on the math), skip it — the graph shows the typical dependency order, not a mandatory gate.
The Full Curriculum
Systems, Python, and math fundamentals — before any AI-specific content.
Classical ML through modern LLMs — every model family covered in depth.
Agentic systems, and where AI meets specific domains — science, healthcare, and beyond.
Designing, building, and running production ML/AI systems at scale.
Knowing whether a system is good, secure, and safe — not just whether it runs.
Reading the literature, and building real things — from scratch, and as full projects.
Tying everything together into a learning path and interview readiness.