Episode 3 of the podcast landed Wednesday, and it is the most personal one yet: the moments that pushed me out of the system after 18 years in public education. This issue pairs it with the practical flip side, a 30-minute cadence review that catches the invisible workload before it breaks anyone on your team, plus three reads worth your time.
Out Loud: Two Minutes vs. An Hour
Episode 3 of Evalve Out Loud is live: What Broke Me. Listen on Spotify or Apple Podcasts, or watch on YouTube. Fair warning, this one is heavier than the first two. It's about the things that pushed me out of the system. But buried in the middle of it is the most practical story I've told on the show so far.
I automated an email. Various organizations send out free virtual professional learning opportunities, and they pile up in an inbox where nobody benefits from them. The old way: spend an hour-plus combing through messages, copying dates and times, organizing sessions by topic, formatting something readable. The new way: a prompt that collects them, organizes the learning by topic and area of focus, and drafts the email. Under two minutes from prompt to send.
Within 24 hours I had a stack of thank-yous. Same information, same inbox it came from. The only difference is that a system did the assembly instead of a person.
That's the line I keep coming back to in this episode. In education, the "system" is usually a human doing manual work, and when that human runs out of hours, we call it a dedication problem. It's not. If the job requires work outside working hours to get done, that's not on the person. That's a broken system.
The fix doesn't start with an expensive platform. It starts with one repeating task you do by hand, and a prompt that does the assembly for you.
Worth Your Time
Introducing Claude Corps by Anthropic. A new national fellowship for people early in their careers who want to bring AI to communities across the country. If you know a recent grad or a career-changer who is serious about doing this work in education, this is a real on-ramp, not a webinar. Applications close mid-July for the first cohort.
Legislative Tracker: 2026 State AI in Education Bills by FutureEd. A running list of every state AI-in-education bill moving this session. I keep a tab on this because it is the fastest way to see what is coming at your district before it shows up as a mandate. Five minutes here beats a surprise in August.
How Schools Are Using AI in 2026: 7 Practical Use Cases From School Leaders by Third Space Learning. Seven concrete use cases from leaders who are already running them, not vendor demos. Read it looking for one idea you can copy this week, not a strategy. That is the right way to use a list like this.
Slow billing compounds fast. The Tabs Billing Lag Calculator shows SaaS finance teams exactly what it's costing them — in two minutes.
Try This: run a 30-minute AI cadence review
The Cadence phase, Days 31 to 60 in my 30/60/90 AI Implementation Plan, is where most AI work either builds a habit or quietly stalls. The fix is not a bigger tool. It is a short, repeating review. Here is one you can run in 30 minutes with whoever is closest to the workflow.
Block 30 minutes. Bring one workflow you started in the last month, the person who uses it day to day, and the exit criteria you wrote on day 1. If you did not write exit criteria, that is your first finding.
Spend the first 10 minutes on one question: is this saving time or costing it? Be specific. Not "it's going well." Closer to "it cut the first draft from 40 minutes to 10, and the person still edits every one." Write the real number, even a rough one.
Spend the next 10 minutes on trust. Ask the person using it where they stopped trusting the output and started double-checking everything. That spot is where your next fix lives. Usually it is one step, not the whole tool.
Spend the last 10 minutes deciding the next two weeks. One of three things: keep it as is, change one thing, or stop. Name it out loud and write it down. Put the next review on the calendar before you leave the room.
Run that every two weeks through the Cadence phase. By day 60 you will know, with evidence, whether this workflow is heading for "scale it" at day 90 or for an honest "cut it." Either answer is a win, because both beat drift.
That is the whole tip. A recurring 30 minutes, three honest questions, one decision. The cadence is the product.
If you would rather run this from a fill-in document than a habit you have to remember, I turned the whole pilot arc into two nine-dollar tools: The AI Pilot Plan Builder and The Day-90 Keep/Kill/Scale Scorecard.
That's it for this one. If something here landed, share it with someone who'd get something out of it. If it didn't, that's part of the deal too. You can find more of what I'm working on at evalveconsulting.com, or book a call if you want to talk through what you're dealing with.
Talk soon,
Chris
Keep up with AI in 5 minutes a day. MavSource aggregates updates from all major AI newsletters, podcasts, companies, AI labs, and hundreds of other sources — then summarizes the key updates.