Free course for software engineers
From software engineer to AI engineer
Six trails that teach what breaks LLM and agent systems in production, and how to design them so they hold up: evals, RAG, agents, security, and cost. Short video lessons, real code, and a graded project at the end of each trail, with a verifiable certificate.
- Free
- No sign-up
- English and Portuguese
- Open example code
Trails
Pick a trail
Six trails, one course. Each builds on the one before, and no lesson repeats. New to LLMs? Start with the first.
- 01Start hereLLM foundations for engineersStart here. What changes when a model becomes part of your system: prompts as interfaces, structured output, tools, and when a fixed workflow beats an agent.Final project: Store support on WhatsApp
- 02Evals and reliabilityHow to know an LLM feature works before your users tell you: evals, golden datasets, structured contracts, and deterministic work moved back into code.Final project: Invoices into the ERP
- 03RAG and contextGive the model the right information at the right time: embeddings, retrieval, hybrid search, reranking, chunking, and long conversations that do not rot.Final project: Internal knowledge chat
- 04Building agentsGive the model autonomy without the classic accidents: loops with a budget, a small toolset, subagents, and multiple agents only when they pay off.Final project: Coding agent
- 05Securing AI systemsFor whoever answers for what the agent does with money, data, and customers: prompt injection, least privilege, human approval, and an audit trail.Final project: Finance assistant with approval
- 06Production: cost, latency, and operationsCheap, fast, and observable: tracing, parallel steps, caching, model cascades, a fallback when the provider goes down, and a stack without vendor lock-in.Final project: Deep research with citations
How every trail works
- 01
Watch
Every lesson opens with a short video. English narration, Portuguese subtitles.
- 02
Read and check
Right below the video is the material: diagrams, code, and quick questions to check what stuck.
- 03
Build
The trail ends in a project: you design the architecture of a real system and the tests tell you if it holds up.
- 04
Prove it
Pass the project and your certificate is issued on the spot, signed and verifiable, ready for LinkedIn.
Interview prep
AI engineer interview prep
The topics that come up in AI engineering interviews and are not in a trail yet: failure signs, refactorings, design patterns, and system design cases. Use it as a review before the technical round.
- 18 topics
Spot the problem
The interviewer describes a system and asks what will go wrong.
- 18 topics
Fix it without breaking it
How to change an LLM system in production, with an eval before and after.
- 47 topics
Design patterns
The toolbox for designing LLM apps and agents.
- 18 topics
AI system design
Full cases, from the naive prototype to an architecture that holds up with real users.
Your next role is AI engineer
Start with LLM foundations for engineers: 14 lessons in order, with progress saved in your browser.