For Students & Developers

Resources for building with AI, not just using it

Instructor-curated tools, articles, and frameworks. No hype, no sales pressure — just what's actually useful when you're building real AI projects.

AI Coding Tools

These are the tools worth learning. Each entry includes what it's actually for, when to reach for it, and the mistake most beginners make.

Cursor

AI-first code editor built for pair-programming with LLMs. Understands your codebase and assists inline.

Free

When to use

When building features, refactoring, or navigating a large codebase with AI suggestions inline.

Common mistake

Using it as autocomplete instead of a thinking partner. Ask it to explain tradeoffs, not just generate code.

Claude

Conversational AI assistant with strong reasoning, code generation, and long-context understanding.

Free

When to use

Architecture decisions, code review, explaining complex systems, and working through design tradeoffs.

Common mistake

Accepting the first output without verifying logic. Always ask it to justify its approach.

GitHub Copilot

In-editor AI code completion trained on public repositories. Integrates directly into VS Code and JetBrains.

When to use

Boilerplate generation, writing test cases, and filling in familiar patterns you would write yourself.

Common mistake

Trusting completions in security-sensitive or data-handling code without review. Copilot doesn't know your context.

Codeium

Free AI code completion and chat assistant that works across most editors and IDEs.

Free

When to use

When budget is a constraint but you still want in-editor AI assistance during development.

Common mistake

Treating it as equivalent to Copilot. Its context window is shorter, so complex multi-file tasks need more guidance.

Learning Resources

How to Think About AI Systems (Not Just AI Tools)

A practical introduction to the difference between using AI features and designing AI workflows.

Prompt Engineering vs. System Design

Why better prompts are not the same as better systems, and what that means for your projects.

Frameworks & Approaches

Structured ways to think about AI-assisted development that hold up beyond tutorials.

Workflow-First Design

Define the steps, decisions, and data flows before writing a single prompt. Prevents brittle systems.

Structured Output Validation

Techniques for making LLM outputs reliable in production: schema validation, retry logic, and fallback paths.

Want to apply these to a real project?

There are optional project-based sessions available for students who want to go beyond tools and frameworks — and actually build something they can explain and show. No pressure, and it starts free.

See the AI Project Sprint

Stay updated

Get notified when new resources or sessions are added. No spam, unsubscribe any time.