Six Week AI
Everyone has an opinion about what AI should do next. Almost no one can tell a good AI use case from a bad one.
A one-page test for any workflow: is it a real AI win, or a money pit? Decide go / not-yet / no, and be able to say why, in plain language, in the meeting.
One email to confirm, then the PDF is on its way. No drip sequence.
Stop guessing in the meeting. When someone says “can’t we use AI for this?”, you’ll have a clear, repeatable way to answer, in an afternoon, not a quarter.
Tell the real wins from the impressive-sounding ones. Score any workflow on whether AI is the actual unlock, or whether you could’ve scripted it years ago.
Back your call with a reason. Not “it feels risky”: the specific gate it fails (data, accuracy, judgment, or compliance) and what would unblock it.
A glance at what's inside
One email to confirm, then the PDF is on its way. No drip sequence.
A method by Ajit Krishna, 20 years across product, design, and engineering; most recently ServiceNow, Reddit, GOV.UK.
Want it applied to your actual workflows? The 90-minute workshop ($2,500, standalone) →