Free Lesson

Make Your AI Agent Think Critically: 3 md files + sub-agent

Part of The AI Evaluation Handbook

Hosted by Vicky Zhao

Wed, Jul 8, 2026

9:30 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

Make Your AI Agent Think Critically: 3 Markdown Files + 1 Sub-Agent
Vicky Zhao
View syllabus

What you'll learn

3 markdown files that give your AI agent better judgement

You'll create a simple 3-file system that gives your AI agent clear rules for better thinking and execution.

Spawn a reviewer sub-agent to notice & fix output issues

You'll add a reusable reviewer that audits outputs against your rules and catches weak evidence or skipped assumptions.

Turn your judgment into reusable frameworks

You’ll build a system for stronger memos, analysis, writing, decisions, and quality-critical workflows.

Why this topic matters

Your AI doesn't need better vibes. It needs better judgment. Most professionals use AI to produce more output. But leaders do not need more polished slop. You need AI output that can follow explicit standards, check its assumptions, think critically, and self-correct when it has drifted from the rules you gave it. We will build a simple critical-thinking layer for your AI agent using 3 markdown.

You'll learn from

Vicky Zhao

Vicky Zhao teaches 295K+ non-technical experts, operators, and leaders how to build AI systems that run on their judgment and frameworks. She uses Claude Code and Codex to manage real workflows across writing, research, business operations, knowledge management, and decision support. Turn your expertise into reusable AI context, frameworks and agentic workflows, so you get real work down without redoing what AI produces.

Previously at Google, L.E.K. Consulting

Google
L.E.K. Consulting
YouTube
See all products from Vicky

Sign up to join this lesson

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