I didn’t expect a performance review to become a podcast episode. But that’s exactly what happened when I sat down to write my annual self-evaluation, asked AI to take a look at it, and AI told me — in no uncertain terms — that I wrote my performance review like I did some stuff and nobody died.
That stung. And it was completely accurate.
This episode is the story of what happened when I stopped treating my self-evaluation as a documentation chore and started using AI as a genuine research partner — going back through my status reports, calendar, and sent emails to surface a full year’s worth of work I had quietly minimized into nothing.
How it started: the dishwasher moment
I wrote my initial self-evaluation the way I always have: from memory, looking at my calendar, summarizing what I could recall. When I handed it to ChatGPT and asked it to compare my original draft against everything it found in my files, the feedback was blunt. I wasn’t describing my work accurately. The problem, it said, wasn’t confidence — it was scale. I had compressed year-long initiatives, cross-functional coordination, and a major integration project into language that communicated nothing about their size or impact. I described my work the way you’d describe unloading a dishwasher.
The four-step process
Instead of starting from scratch, I asked Copilot — connected to my Microsoft files — to work through three separate sources one at a time: my weekly status reports to my supervisor, my calendar patterns (looking for projects behind the meetings, not summaries of individual ones), and my sent emails (looking for follow-ups, deliverables, and places where I helped colleagues get things done). Each prompt produced a possible accomplishment list. Then I gave it the broad categories from my actual job description and asked it to synthesize a new draft. The result surfaced projects I’d genuinely forgotten, captured the scope of things I’d filed under “just part of the job,” and connected dots I hadn’t connected myself.
What AI saw that I couldn’t
When it handed back the new draft, my first reaction was: this is too much, this is bragging. But when I went back through and checked each item against actual evidence — it was right. I had led a major integration project. I had coordinated across multiple business and technical teams. I had been the point person and subject matter expert throughout. AI wasn’t inflating anything. It was using language that already existed in my own record and organizing it into something coherent.
Humility vs. inaccuracy
This is the shift that mattered most to me. Humility says: “I know I didn’t do this alone. I have more to learn.” Inaccuracy says: “I barely did anything.” I had been writing the second one while thinking it was the first. The performance review is not the place to pretend things didn’t happen. It’s where you give your manager an accurate story they can take into the room where your future gets discussed.
What I shared with my team
After finishing my own review, I wrote up the full prompt set and shared it with my team — not to show off, but because if AI can do this for me it can do it for anyone. Clean, repeatable, something anyone can run in minutes. I stuck my neck out admitting I’d used AI for this. I think it was worth it.
Your small step
Think of one area — work, a creative project, something at home — where you’ve been doing consistent, sustained effort you haven’t fully acknowledged. Pull up the evidence: emails, notes, outcomes. Ask AI: What do you see here? What patterns do you notice? Don’t start with “help me sound impressive.” Start with “here’s what actually happened. Help me see it more clearly.” You might discover the thing you’ve been calling “I did some stuff” is a much bigger story than you’ve been telling.
Jill’s Links
http://jillfromthenorthwoods.com
https://www.youtube.com/@startwithsmallsteps
https://www.buymeacoffee.com/startwithsmallsteps
https://twitter.com/schmern
Email the podcast at [email protected]
By choosing to watch this video or listen to this podcast, you acknowledge that you are doing so of your own free will. The content shared here reflects personal experiences and opinions and is intended for informational and educational purposes only. I am not a software developer, data scientist, or AI professional. Any tips, tools, or suggestions offered should not be considered a substitute for professional technical advice. AI tools and platforms change frequently — always verify current features, pricing, and terms directly with the providers. You are solely responsible for any decisions or actions you take based on this content.
