Building an LLM Council
Put your questions to a council of different AI models.
$ open "Meet the Council"
Start your first council run and watch three models answer, rank, and synthesize.
checkpoints run as you work
✓ checkpoint passed, lab 2 unlocked
What you'll build
Curriculum
8 labs, about 2 h
- 1
Meet the Council
Start your first council run and watch three models answer, rank, and synthesize.
14 min
Pro - 2
Read the Responses
Compare the independent answers and pull out the real disagreements.
10 min
Pro - 3
Inspect the Ranking
See how blind ranking works and reveal which model wrote each answer.
10 min
Pro - 4
Choose a Roster
Rerun the same question with the fast preset and compare the two councils.
15 min
Pro - 5
Judge a Product Idea
Use the council as an advisory board to score and pressure-test a startup pitch.
15 min
Pro - 6
Review a Piece of Writing
Turn the council into an editorial panel and produce an ordered edit plan.
15 min
Pro - 7
Drive a Decision
Get a decision memo with a recommendation, a comparison, and the minority view.
15 min
Pro - 8
Ship Your Council
Run the council on your own question and ship a shareable HTML report.
18 min
Pro
Your instructor
Elvis Saravia
Founder, DAIR.AI
Elvis founded DAIR.AI and wrote the Prompt Engineering Guide, one of the most widely used references on working with language models. He has worked at companies like Meta AI and Elastic, and specializes in AI agents, RAG, and harness engineering.
How labs work
Live workspace
You run the real tools in a browser workspace. Nothing to install.
Checkpoints
Each lab checks your files and output, then unlocks the next one.
Your work stays
The workspace persists between labs so you finish with a project.