AIF1RSTAcademy
All missions
AI for WorkBuildSupport level: Guided

Your First AI Coworker

Design one AI Coworker for a repetitive part of your work: brief it, test it on real cases, and document how to supervise it.

Campaign:
Launch Your First AI Coworker
Estimated effort:
~180 minutes total

The problem

Part of your job repeats every week — drafting, summarizing, classifying, preparing — and it competes with the work that actually needs you.

Why it matters

This is the unit of every AI-powered team: one person, one well-briefed AI Coworker, clear supervision. Learn it on your own work first.

Context

You know your role better than any tool does. This mission turns that knowledge into a working brief for an AI Coworker that assists one specific task, with you supervising every output.

Objective

Produce a documented, tested AI Coworker brief for one repetitive work task, validated on three real cases.

Pain point

Repetitive work that consumes focus hours and gets done tired, late, or inconsistently.

Opportunity

A supervised AI Coworker gives you back focus time and produces more consistent first drafts than a rushed human ever does.

Expected transformation

From doing a repetitive task alone to supervising an AI Coworker that produces the first pass — documented well enough that a colleague could use it.

Prerequisites

None — this is a first mission.

Risks and guardrails

  • Pasting confidential work data (clients, employees, internal figures) into external tools.

    Guardrail: Check your employer's AI policy first. Use anonymized or synthetic examples in the brief and tests.

  • Passing AI output onward without review.

    Guardrail: The brief defines a review step: you check every output against the quality bar before it leaves your desk.

  • Picking a task with judgment calls the AI shouldn't make.

    Guardrail: Checkpoint 1 filters for tasks with clear inputs and checkable outputs — decisions stay with you.

Checkpoints

  1. 1

    Pick the right task

    List your repetitive tasks. Pick one with clear inputs, a checkable output, and no confidential data required. Write down what 'good' looks like for it.

    ~40 min

  2. 2

    Write the Coworker brief

    Write the role, inputs, steps, tone, limits, and quality bar as instructions. Include two worked examples (anonymized).

    ~60 min

  3. 3

    Test on three real cases

    Run the brief on three real (anonymized) cases. Score each output against your quality bar, fix the brief where it failed, and note the corrections.

    ~80 min

Deliverables

  • The AI Coworker brief — A reusable instruction document: role, inputs, steps, limits, quality bar, and worked examples.
  • Test log — Three real cases with outputs, your review notes, and the brief corrections they triggered.

Evidence

  • The brief plus the test log, anonymized — no client, employer, or personal data.

Evidence is private by default. You choose if it ever becomes anything else — and that choice requires your explicit consent.

Success criteria

  • The brief ran on three real cases and produced usable first drafts.
  • Every output was reviewed against an explicit quality bar.
  • At least one brief correction came out of testing — documented.
  • No confidential data appears in the brief, tests, or evidence.

Review

Self-review with rubric · Peer review

Self-review against the success criteria is available today. Reviewed verification beyond self-review opens as the Founding Community's review circles form.

Reflection

  • Which review step can never be skipped for this task, and why?
  • What did testing change about your original brief?
  • What would you need before trusting this Coworker with a bigger share of the task?

Impact measurement

Baseline: minutes per instance done by hand. Result: minutes per instance with the Coworker (including your review). Category: time saved / workflow improved (self-reported).

Capability rewards

  • Context design · 40 XP
  • Prompting · 40 XP
  • Evaluation · 30 XP

What this unlocks

  • AvailableThe brief format as a template for your next AI Coworker.
  • In preparationThe next mission in Launch Your First AI Coworker: measuring a month of supervised operation.

Privacy

Everything is private by default. Anonymize all work data — the mission is designed to be completed without sharing anything confidential.

Accessibility

Self-paced, text-based, no timers. The brief can be written in any tool you already use.

Need help? The help ladder

Help arrives in graduated steps so the work stays yours — open only as far as you need.

  1. 1Objective reminder

    Produce a documented, tested AI Coworker brief for one repetitive work task, validated on three real cases.

  2. 2Guiding question

    Which review step can never be skipped for this task, and why?

  3. 3Hint

    List your repetitive tasks. Pick one with clear inputs, a checkable output, and no confidential data required. Write down what 'good' looks like for it.

  4. 4Partial example

    The AI Coworker brief — A reusable instruction document: role, inputs, steps, limits, quality bar, and worked examples. · Test log — Three real cases with outputs, your review notes, and the brief corrections they triggered.

  5. 5Template

    The brief format as a template for your next AI Coworker.

  6. 6Walkthrough

    Work checkpoint by checkpoint in order — each checkpoint description is a step of the walkthrough.

  7. 7AI coach

    Opens with the AIF1RST Guide and community review circles — not yet available.

  8. 8Mentor or staff

    Opens with the AIF1RST Guide and community review circles — not yet available.