AI-901 hands-on

Foundry Lab Hub

Five labs that take you from opening a Foundry project to shipping a grounded, safety-aware assistant. Each lab produces something you can reuse in the hackathon or in second semester project work.

Lab Catalog

Lab 1 30-45 minutes

Foundry Project Setup

Open the assigned Foundry project, verify role access, and confirm the parent resource scope.

Lab 2 1-2 hours

Model Deployment & Calling

Call the deployed gpt-4o-mini or assigned model from Python and from REST, and observe the token cost.

Lab 3 2-3 hours

Single-Agent Prototype

Build a small assistant with clear instructions, a narrow task, and a demo scenario.

Lab 4 Half day

Grounded Assistant (RAG mini-lab)

Upload a small document set, retrieve relevant passages, and produce grounded answers with source references.

Lab 5 1-2 hours

Responsible AI, Cost & Audit

Apply content safety, monitor token usage, and review the audit trail.

Second Semester Module Plan

Week 1

Foundry orientation

Project access, model deployment, first Python call.

Lab page

Week 2

Model call & prompting

System prompts, test cases, hallucination checks, token budget.

Lab page

Week 3

Single-agent prototype

Instructions, tools, demo flow.

Lab page

Week 4

Grounded assistant + sprint

Small RAG, responsible AI review, team prototype sprint.

Lab page

Tips

Run labs in order

Labs build on each other — the project setup unlocks the model call, the model call unlocks the single agent, and so on.

Save artifacts

Each lab tells you what to save — code, screenshots, or a transcript. These become your hackathon evidence and your second semester portfolio.

Use the sample repo

All code in the labs maps to halla-ai/hackathon-sample-2026. Clone it once and edit.

Cost guardrails

Each lab states an expected Azure cost. If your usage exceeds the lab estimate by 3x, stop and ask a tutor.