How to get started in AI as a developer

You don't need a PhD to work with artificial intelligence. Many companies are looking for developers who can build products on top of existing AI models, like Claude.

Two different paths

• Research and model training: math, statistics and a lot of infrastructure.
• Building AI applications: using existing models through their API to solve real problems. This is the most accessible path to start.

What you'll learn

1. Python: the most used language in AI. Variables, functions, lists, dictionaries and installing libraries.
2. Calling a model's API: sending a message and getting a response from your code.
3. Prompt engineering: clear instructions, examples and output format to get reliable results.
4. Embeddings and semantic search: letting the AI answer using your own documents (RAG).
5. Tools and agents: letting the model call your functions, like checking an inventory.
6. Evaluation: measuring whether your application answers well, with test cases.

A good first project

An assistant that answers questions about a document (for example, your university's rules), cites where the answer came from and says "I don't know" when the information isn't there.

Good habits from day one

• Never put your API key in code you push to GitHub.
• Review the answers: models can be confidently wrong.
• Be careful with the personal data you send to the model.

Next step

Codea's free AI Engineer Jr route teaches Python, prompts, APIs, embeddings, tools and evaluation through hands-on projects. Route content is in Spanish.