AI Engineering
In development
From Python for programmers to ML compilers, in five levels.
- 01Beginner
- 02Intermediate
- 03Advanced
- 04Master
- 05Elite
Learn the fundamentals by hand, then catch what an AI assistant gets wrong and ship code that works in production.
An assistant wrote the lookup tool your support agent calls, and its own tests pass. Review it before it goes live.
import sqlite3
def find_orders(db: sqlite3.Connection, name: str) -> list[tuple]:
"""Look up a customer's orders for the support agent."""
query = f"SELECT id, total FROM orders WHERE customer = '{name}'"
return db.execute(query).fetchall()
The assistant's tests
Production checks
Not run yet
Done · Nothing missed
AI writes code in seconds. You still have to know if it's right.
Write the core of each system by hand, until its tests pass.
An AI assistant builds on your code, and its own tests pass. Find and fix what would break in production.
Put your code to work on a real decision. It ships when hidden tests and production checks pass.
Write the core of each system by hand, until its tests pass.
An AI assistant builds on your code, and its own tests pass. Find and fix what would break in production.
Put your code to work on a real decision. It ships when hidden tests and production checks pass.
In development
From Python for programmers to ML compilers, in five levels.
We want to build courses with engineering teams, based on their own systems and standards.