The crew only understands you if you know how to talk to them. Twenty-six rituals across seven units: anatomy and clarity, input shape, examples, asking for the output well, reasoning, debugging, and hardening for production.
YOUR RUN0 / 26
XP BANKED0 / 224
DEBRIEF · WHAT YOU'LL CLEAR
By end of shift, you can
Echo
SIGNS OFF ON THIS UNIT
Read a model's mood from its first sentence
Pick the right role for the right turn
Course path
0 / 26 steps cleared
Continue!
Meet the crew
How to talk to Vega
Vague vs specific: close the doors
The format Echo understands
Write examples that actually teach
Spec a JSON schema Echo will sign off on
Diagnose why a 'good' prompt drifts
Clear a capstone with hearts to spare
System vs user: two channels, not one
Fill in the context: your first introduction
Learn by example: few-shot prompting
How many examples, which examples
Counter-examples: teaching with what NOT to do
The system prompt in depth
Your first prompt: extract data for Echo
Asking for structured output, not just text
When data is missing: explicit nulls
Length constraints: forcing short output
Think before answering: chain of thought
Decomposing a big task
Just answer with X: kill the preamble
Negatives: how to say NO without reinforcing
Iterating on a failing prompt
Prompt failure gallery
Guardrails: keep your prompt from breaking on dirty inputs
When the model refuses: real and false refusals
Ambiguity: when there are two interpretations
Capstone: Atlas signs your first report
Try it yourself
Onboarding complete
Continue!
Meet the crew
How to talk to Vega
Vague vs specific: close the doors
The format Echo understands
System vs user: two channels, not one
Fill in the context: your first introduction
Learn by example: few-shot prompting
How many examples, which examples
Counter-examples: teaching with what NOT to do
The system prompt in depth
Your first prompt: extract data for Echo
Asking for structured output, not just text
When data is missing: explicit nulls
Length constraints: forcing short output
Think before answering: chain of thought
Decomposing a big task
Just answer with X: kill the preamble
Negatives: how to say NO without reinforcing
Iterating on a failing prompt
Prompt failure gallery
Guardrails: keep your prompt from breaking on dirty inputs
When the model refuses: real and false refusals
Ambiguity: when there are two interpretations
Capstone: Atlas signs your first report
Try it yourself
Onboarding complete
0% completed
FIRST OFFICER
What you'll be able to do by the end of this track
