Free Lesson

Build Your Context and Knowledge Layers for Agentic Growth

Hosted by Grace Man

Wed, Oct 7, 2026

4:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

The Agentic Growth OS for Enterprise-Scale Leaders
Grace Man
View syllabus

What you'll learn

Score how far your agent output drifts today

One prompt, three runs, scored against your own positioning, so the drift is visible rather than felt.

Build the context layer and the knowledge layer

What every agent reads first, and the approved truth it draws from, separated so each can change on its own.

Leave with both layers working on your own company

Two working files, the same prompt behaving, and an ordered list of what belongs in the knowledge layer next..

Why this topic matters

Your agents produce good work once and different work the next run. The tone drifts, the positioning is wrong, and every session starts by re-explaining the company. The fix is two layers, not a better prompt: a context layer every agent reads first, and a knowledge layer holding the approved truth it draws from. In 45 minutes you build both in Claude Code on your own company.

You'll learn from

Grace Man

Enterprise ABM | Ex-Microsoft | Founder, AI Strategy League

LinkedIn | Articles | AI Strategy League
Grace has trained and coached more than 3,000 professionals in agentic AI for growth and marketing, with learners from Google, Meta, and more. Grace spent 15 years building and running growth and marketing at Microsoft, Canon, and Pfizer, influencing more than $2B in revenue. She teaches what she runs. AI Strategy League operates on the same agentic system taught in these sessions, and every session builds something live rather than demonstrating a finished result. Keynote speaker and chairperson at the AI and Big Data Expo, DMWF, and Performance Marketing World.

Previously at

Microsoft
Pfizer
Canon
See all products from Grace

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.