AI Learning Paths & Courses
Choose a practical starting point, build a weekly practice rhythm, and learn AI without collecting courses you never finish.
ToolDix AI Learning
A working library for people who want more than a list of links: structured public courses, original study notes, primary readings, video analysis prompts, and small build exercises. Start with the work you want to do, not the tool you happen to see first.
Start here
238 original lessons across 20 paths, written and illustrated by ToolDix. Every path opens to a full outline you can read here.
New lessons are reviewed before publication.
Choose a practical starting point, build a weekly practice rhythm, and learn AI without collecting courses you never finish.
A tool-by-tool guide to the frameworks AI work runs on -- PyTorch, TensorFlow, scikit-learn, OpenCV, LangChain, Ollama and the browser automation stack: what each is genuinely for, when it is the wrong choice, and the failure modes that show up in real projects.
A full deep-dive course on AI agents -- core components, the agent loop, LLM foundations, retrieval, context engineering, and how to build your first agent.
A full deep-dive course on Claude Code -- the permission model, the core plan-edit-test loop, CLAUDE.md project context, subagents, hooks, MCP, skills, and the headless SDK.
A full deep-dive course on OpenAI's Codex -- how it plans and edits code, the CLI, IDE extension, and cloud sandbox modes, task scoping, and reviewing agent-generated diffs safely.
A full deep-dive course on OpenCode -- provider-agnostic model configuration, plan versus build modes, self-hosting, and when an open coding agent is worth the setup over a closed one.
A full deep-dive course on vibe coding -- where it speeds up real work, where it creates real risk, and the review habits that keep a fast, description-driven workflow safe to ship.
Write, test, and improve prompts that give models the context, constraints, and evaluation signals they need.
Learn the model, data, optimization, and evaluation foundations behind modern AI by training and inspecting small systems yourself.
Go from a visual brief to consistent generated images while managing references, revisions, and commercial-use checks.
Plan clips, maintain visual continuity, and combine generation with editing rather than relying on one-shot video prompts.
Use AI for music ideation, arrangement, and iteration while keeping licensing and voice rights in view.
Learn text-to-speech, avatars, lip sync, consent, and disclosure practices for trustworthy synthetic media.
Explore text-to-3D and image-to-3D workflows, then prepare assets for real editing, rendering, or game pipelines.
Build node-based image workflows that are reproducible, portable, and easier to debug than a growing pile of screenshots.
Understand transformer-based language systems, build retrieval with citations, and evaluate quality before treating a demo as a product.
Move models from notebooks into reproducible services with versioned data, automated tests, deployment controls, and production monitoring.
Choose valuable AI problems, design trustworthy human-AI interactions, and connect experiments to measurable product and business outcomes.
Turn responsible-AI principles into risk registers, threat models, evaluation gates, incident plans, and accountable operating practices.
Use AI to support teaching and learning with clear goals, privacy safeguards, and academic-integrity boundaries.
Featured resources
Start from a focused selection, then browse the complete library by format. Every card opens a ToolDix guide with source context and study notes.

DeepLearning.AI · Andrew Ng
A non-technical orientation to what generative AI can do, where it fails, and how teams can use it responsibly.

Microsoft
A structured, code-forward curriculum covering generative AI concepts and application building.

Hugging Face
A practical path through Transformers, tokenizers, datasets, fine-tuning, and modern LLM workflows.

Hugging Face
An applied route through agent concepts, tool use, and agent frameworks.

OpenAI
Official documentation for delegating coding tasks, providing context, reviewing changes, and using Codex within an engineering workflow.

Full Stack Deep Learning
A practitioner-focused course covering prompt engineering, augmented language models, LLMOps, deployment, user experience, and product development.