AI Thinking Partner: The System Behind My Decisions

A CMO’s job is running systems. I already had mine: structured inbox, clear priorities, regular follow-ups, frameworks I use for decisions.

But there’s a gap between knowing a framework and actually applying it when you’re in your 10th meeting of the day. Between promising to follow up and actually doing it on time. Between reading a book on execution and using it three months later.

So I built an AI assistant that closes that gap.

It knows the thinking models I use — 4 Disciplines of Execution, Dorofeev’s methodology, risk analysis. When I’m making a decision, it applies the right framework — even when I’ve forgotten it exists in the heat of the moment.

After meetings, it parses transcripts, logs commitments, syncs them to Asana, and reminds me to check. Follow-through improved — because the system doesn’t forget.

I talk to it in Telegram via voice messages. It sorts inputs into tasks, emails, ideas. I can draft a reply in seconds without opening Gmail.

The whole thing is built with Claude Code and Python. No engineering team. One person, vibe coding.

This isn’t about replacing a manager’s thinking. It’s about making sure good systems actually run consistently — not just on your best days.

The most valuable part? The thinking partner. It interviews me before I make a decision. Runs premortems. Challenges my assumptions. Catches me when I’m rushing.