Capstone challenge
Ship It
You're the PM launching Initech's AI support bot.
Pick a setup, pass the launch review on real recorded runs, then ship it for less.
Your launch setup
Defenses
0 reviews so far
Need a refresher?
- Evaluating AI Changes
An eval is a fixed test set plus graders, giving a score you can compare across changes.
Open the lesson(opens in a new tab) - Writing Good Prompts
A prompt is a set of ingredients, and you can measure each one. Spelling out the format, defining your categories, and covering edge cases usually matter far more than telling the model who it is.
Open the lesson(opens in a new tab) - Retrieval-Augmented Generation
RAG = search your own data first, then let the model answer from it.
Open the lesson(opens in a new tab) - Prompt Injection & Guardrails
Prompt injection hides instructions inside content the model reads. Check what comes back from tools, not just what the user types, and limit what the agent is allowed to do.
Open the lesson(opens in a new tab) - Human-in-the-Loop
Letting an agent act on its own is fast until it refunds the wrong customer. Asking a person every time is safe but slow. Route only risky actions (money, anything irreversible) to a person, and keep the policy in the agent's prompt so less reaches them at all.
Open the lesson(opens in a new tab)
Launch review
Choose a setup and run the review: it replays real recorded runs of each part.
Leaderboard
Loading…