Hands-on, self-paced coding labs
81 labs and 1773 chapters in Python, AI engineering, JavaScript and TypeScript, Java, Go, Rust, SQL and Postgres, data science and DevOps. Each lab follows one story, a small company with a real problem, and you build its code one chapter at a time.
See all labs · Learning paths · DSA, LLD and CS fundamentals
How a lab works
- Read a short lesson: why the idea matters, then how it works, with animations.
- Write the code in a real editor and terminal in your browser. Nothing to install.
- Press Check. Tests run on your code and tell you, task by task, what works and what to fix.
AI suggestions and a tutor sit in the editor to explain, not to hand over answers. Easy mode shows a solution when you are stuck; Hard mode keeps it locked for more points and a Hard mode certificate.
Labs by topic
AI engineering
Models, RAG, agents, MCP, evals, memory, and the frameworks that build them.
- LangChain and LangGraph
- Pydantic AI
- OpenAI Agents SDK
- Building with LLMs
- Building with LLMs (TypeScript)
- RAG in Python
- Agents from scratch (TypeScript)
- Harness engineering
- Harness engineering (TypeScript)
- Build a voice agent
- Build a voice agent (TypeScript)
- Fine-tuning basics
- Fine-tuning basics (TypeScript)
- Build an AI gateway
- Build an AI gateway (TypeScript)
- Claude and the Anthropic SDK
- Claude and the Anthropic SDK (TypeScript)
- The Vercel AI SDK
- Build your own coding agent
- Build your own coding agent (TypeScript)
- AI security: attack and defend
- AI security: attack and defend (TypeScript)
- Agents from scratch
- MCP servers and clients
- MCP servers and clients in TypeScript
- Evals that catch regressions
- Evals that catch regressions in TypeScript
- Memory and multi-agent systems
- Memory and multi-agent systems in TypeScript
- RAG: search that answers
Projects
One real codebase built chapter by chapter, yours to push to GitHub at the end.
- Booking backend project
- Ordering backend project
- AI support copilot project
- AI ordering agent project
- AI invoice reader project
- MCP server project
- AI answer grader project
- Build your own GPT project
- AI support copilot project (TypeScript)
- AI booking agent project (TypeScript)
- AI invoice reader project (TypeScript)
- MCP server project (TypeScript)
- AI answer grader project (TypeScript)
- Build your own GPT project (TypeScript)
Interview bug cases
Tickets from production: reproduce the bug, fix it, guard it with a test, and explain it as you would in an interview.
- Python traps
- Git disasters
- Concurrency and races
- SQL traps
- Node.js traps
- React traps
- AI engineering traps
- AI engineering traps (TypeScript)
- TypeScript traps
- Go traps
- Python traps 2
- Java traps
- JavaScript traps
Python
Python from the first print to production APIs.
JavaScript and TypeScript
The web from HTML and CSS to JavaScript, React, Node.js and APIs with Express.
- HTML and CSS
- React
- JavaScript, inside out
- TypeScript for real code
- How Node really works
- APIs with Express
Java
Core Java and Spring Boot services.
Go
Go from the first program to concurrent services with HTTP and Postgres.
Rust
Rust from the first program to ownership, traits, threads, async with tokio and a TCP service.
Data science
NumPy, pandas, Matplotlib and machine learning, each with a real company and its data.
Databases
SQL, Postgres and Redis, from the first query to production.
DevOps and tools
Git, Linux, Docker and Kubernetes: from the tools every developer uses daily to running services on a cluster.
Learning paths
- AI engineer with Python: Build LLM apps, RAG, agents and MCP servers in Python, then measure them.
- AI engineer with TypeScript: The same AI engineering, in the language of the web.
- Agent engineer: Build agents that act on their own, safely, and prove they work.
- AI platform engineer (LLMOps): Run AI in production: cost, uptime and safety, at scale.
- AI product engineer with TypeScript: Add AI to real apps, front to back, in one language.
- Forward deployed engineer: Take AI into a customer’s world: their data, their APIs, their rules.
- Backend developer with Node.js: APIs, databases and caching in JavaScript and TypeScript.
- Backend developer with Python: APIs, databases and caching with Python and FastAPI.
- Backend developer with Java: Core Java and Spring Boot services on Postgres.
- Backend developer with Go: Fast, simple services in Go.
- Front-end developer: Pages, components and state, from HTML to React.
- Full-stack developer: React in front, Node.js and Postgres behind, one language.
- Database master: From your first SELECT to reading query plans.
- DevOps engineer: Linux, Git, containers and Kubernetes, the way on-call needs them.
- Data scientist: From arrays to models, in Python.
- Model builder: How models work inside, from arrays to a GPT you train yourself.
Questions
Is it free to try?
Yes. The first 2 chapters of every lab are free with a Google or GitHub sign-in.
What do I need?
A browser. Every lab runs its own container with the language, the database and the tools already set up.
Do I get a certificate?
Yes, when you finish every chapter of a lab, with your own name on it and a link anyone can check.