Free practical AI courses
Practical routes for learning Codex, local AI, agents, and real production workflows. No account required: Aulafy does not create an account or use personal tracking cookies. Learning progress stays in this browser.
Choose a task, not a technology
Open one concrete outcome. We will introduce the course, tool, or concept only when you need it.
Try AI for the first time
I have never used it or do not know what to ask.
One small request, reviewed and under your control.Improve an email without inventing facts
I want to write faster while reviewing before I send.
A bounded draft with visible gaps and human review.Study without copying the answer
I need to understand, practise, and check what I remember.
A tutor method with questions and unaided recall.Detect when AI made something up
The answer sounds convincing, but I do not know whether to trust it.
A claim classified as confirmed, uncertain, or rejected.Prepare a small-business AI pilot
I want to apply AI without touching critical processes yet.
A bounded, reversible case with explicit stop criteria.Build a first local website
I have an idea and want to see it working before publishing it.
A responsive page you can inspect on your computer.Fix an Ollama error
Ollama will not connect, closes, or does not respond as expected.
An ordered diagnosis before you reinstall or change everything.Turn a statement into CSV with control
I need structured data without importing anything automatically.
A reviewable, reconciled CSV that leaves accounting untouched.Compare tools without rankings
I do not know whether to choose ChatGPT, Claude, Gemini, or something else.
The same test for every tool and a documented decision.DeepSeek Harness (DSH), from zero to expert
Understand DeepSeek's agent harness, install DSH safely, configure DeepSeek models, and design auditable workflows with modes, plugins, workspaces, subagents, metrics, and profiles.
You finish with: A practical project with explicit verification and next steps.
Aulafy foundations
Prepare a local-first environment for any Aulafy course: terminal, modern Python with uv, Git, AGENTS.md, Docker, local services, and a reproducible project template.
You finish with: An Aulafy base project with README, AGENTS.md, Python environment, clean Git history, and services ready.
Codex for programmers
Learn a professional Codex workflow: explore repositories, write verifiable requests, configure AGENTS.md, implement with tests, review Git changes, control permissions, and automate tasks with codex exec.
You finish with: A complete repository change with a test, review, and clean commit.
AI Router and content system
Design an AI router that chooses between local models and frontier APIs with LiteLLM, privacy policies, quality scoring, shadow mode, observability, and human review for educational content.
You finish with: An AI Router prototype for generating and reviewing educational content with auditable routes.
Claude Code, from zero to pro
A practical route from installation to skills, subagents, MCP and professional workflows, including local AI apps that run on your own computer.
You finish with: An AI-assisted development workflow adapted to your work.
Claude Code + Local AI
Learn to connect Claude Code with Ollama, local models, vector databases and desktop workflows you can run without sending every task to the cloud.
You finish with: A working local AI application using RAG, voice, PDF, or productivity tools.
Generative AI: image, voice and video
A hands-on course for images, audio, video and creative automation, with attention to quality, control and repeatable workflows.
You finish with: A practical project with explicit verification and next steps.
Games, 3D and simulations with AI
Build playable prototypes, 3D scenes, mechanics and asset workflows using AI as a practical production partner.
You finish with: A practical project with explicit verification and next steps.
AI security and evaluation
A practical security path for prompts, agents and models: evals, red teaming, data risks, guardrails and responsible deployment.
You finish with: A practical project with explicit verification and next steps.
Local MLOps and model deployment
Move from local experiments to controlled deployments with model serving, logs, evaluation loops and operational discipline.
You finish with: A practical project with explicit verification and next steps.
Fine-tuning and post-training for LLMs
Prepare datasets, run LoRA and QLoRA jobs, evaluate overfitting and export models for local use.
You finish with: A practical project with explicit verification and next steps.
Agents and automation
Learn subagents, tools, memory, MCP, retries and automation loops for real workflows on your own machine.
You finish with: A practical project with explicit verification and next steps.
Production agents with LangGraph and n8n
Compare frameworks, design state and memory, add human approvals, observe failures and ship agent workflows responsibly.
You finish with: A practical project with explicit verification and next steps.
Self-hosted AI automation for small businesses
Deploy n8n, Ollama, Qdrant and support automations with Docker, VPS setups, queues, backups and monitoring.
You finish with: A practical project with explicit verification and next steps.
Advanced and secure RAG
Go beyond basic embeddings with better chunking, reranking, permissions, citations, monitoring and production safety.
You finish with: A practical project with explicit verification and next steps.
AI for small businesses and freelancers
Identify valuable use cases, automate office flows, build useful assistants and decide what should stay human.
You finish with: A practical project with explicit verification and next steps.
Coming soon
AI for data and analytics
CSV, SQL, notebooks, dashboards, data cleaning and analytical assistants with open models.