Skip to main content

We earn commissions, at no additional cost to you, when you shop through the links below. Learn more

DevBench

AI Engineer roadmap

Role-based

A practical path to becoming an AI engineer: the Python and API basics, how LLMs work, prompting and structured output, retrieval, tool use and agents, and the evaluation and safety habits that make AI features reliable in production.

24 topics
0%

Progress is saved in this browser only — no account needed.

  1. 1Foundations

    The programming and ML ideas everything else sits on.

  2. 2How LLMs work

    You don't need to train one, but you do need a correct mental model.

  3. 3Building with LLM APIs

    Turn a model into a dependable component of a product.

  4. 4Retrieval and your own data

    Give the model facts it was never trained on.

  5. 5Tools and agents

    Let models take actions, carefully.

  6. 6Evaluation, safety and production

    What separates a demo from a feature people can rely on.