Chris Pallé
↑Overview

03 · World Vision USA · Nonprofit · 2023–2025

Harnessing off-the-shelf GenAI

Problem
Every good AI result was a one-off nobody could repeat.
Action
Tested what off-the-shelf tools could really do, then built the context they lacked: specs, examples, critique.
Solution
A 17-person team using GenAI that holds up in anyone’s hands.
  • 4–5More runs of the spec, then a write-up for other teams
  • On stageThe artist-appeal backdrop
  • Snap ShotsThe war room, still running
Off-the-shelf AI tried broadly; the frame that held gets a harness of context; the spec runs four or five more times, then goes to other teams as a write-up. Below, the loop the war room ran: try, critique, write it down, run again, still running as Snap Shots.
The full story · 2 chapters

The full story, chapter by chapter

I · The full story

Being written now, a chapter at a time.

  1. IWhat changed for the teamSame tools for everyone. What changed was who could repeat the result.
  2. IIThe situationI was hired to lead design. The job became two jobs.
  3. IIIFour decisionsAnd what I decided against.
  4. ◆Feature · the UX war roomThe UX war room.
  5. IVWhat it looked likeThe work, as you would have seen it.
  6. VWhat I’d do differentlyA design system needs a dedicated advocate.
  7. VIWhat this showsTwo things a design or product lead can use.
  8. VIIProofProof
  9. VIIIWorking with meFor a team past the first good AI result that needs the second.

Chapters arrive one at a time. Hear the whole story on a call

II · For your team

You don’t need a custom model to get dependable AI work. You need the right context around the tools you already pay for.

Working on something like this?