Mond(AI)y Coffee

What are we actually getting back from enterprise AI?

Pre-read

Chikamso Anayor Achu

Meeting: November 2, 2026
Jump to a section

Mond(ai)y Coffee — Speaker Pre-Read

Date: Monday, November 2, 2026
Speaker: Chikamso Anayor Achu — Founder, Agentic Return
Topic: What are we actually getting back from enterprise AI?

Who are you?

What are you building?

I’m building Agentic Return to help companies figure out what they’re spending on AI coding tools like Copilot, Claude Code and Cursor, and what they’re getting back. The idea is to connect that spending to the projects and work it supports, so a team can ask questions about its costs, manage a project’s budget and find ways to get better results from the tools it already uses. I also want agents to share useful project knowledge so companies aren’t paying different agents to keep figuring out the same things. Longer term, that same record of costs and work should help Finance with accounting and tax preparation, with people still reviewing those decisions.

The core idea

Knowing how much you spent on AI isn’t the same as knowing whether it was worth it. I want companies to be able to look at a project and understand what the agents cost, what they helped deliver, and whether it makes sense to spend more or change how the work gets done.

What’s working / not working

  • Working: I’ve built demos. At Microsoft, I helped customers get value from AI coding tools and kept seeing questions around managing the tools, governing their use and connecting the costs to value. That’s what led me here.
  • Not working yet: I’m looking for design partners. I need to test this with real teams and budgets to learn which parts are genuinely useful and where the product needs to change.

Open questions

  • How are you measuring whether AI is actually paying off in your organization today?
  • What would make you confident enough to spend more on an agent or workflow? What would make you pull back?
  • How do you keep control of AI costs without discouraging experimentation or slowing people down?

Spicy take

I think once everyone has access to pretty much the same AI intelligence, the biggest difference is going to be how fast you can move. And a lot of that comes down to capital and token economics—how much AI you can afford to use, and what you can get done with it.

If we only discuss one thing, it should be:

The biggest challenge with enterprise AI adoption is figuring out the ROI—what are we actually getting back from all this spending?

Mond(ai)y Coffee · mondaiycoffee.com

← All meetings