Agentic Analytics 201: Validation & Context Management
Shane Butler
Co-founder @ AI Analyst Lab
Sravya Madipalli
Data Science Leader |Ex-Microsoft
Proven playbook to make your AI data analyst reliable enough to trust at team
You've built an agentic analytics system that works for you. You run Claude Code in the terminal, write skills, chain agents, and ship an analysis end to end. This course is the next step: making that system reliable enough to trust at team scale.
Solo, you can eyeball whether an answer looks right. Across a team, you can't. Two people run the same query and get different numbers. The model your workflow depends on gets deprecated overnight. An AI-written analysis reads beautifully and is quietly wrong. This course engineers past all three.
You'll swap the brain of Claude Code between frontier and open-source models, so you're never locked in or cut off. You'll build a multi-model validation loop where one model builds, another reviews, and you judge the disagreement, plus the four eval types that catch what a single pass misses. And you'll engineer the context and semantic layers that makes any model produce consistent, audience-ready work.
You leave with a complete agentic analytics system you built yourself, one repo that runs on both Claude Code and Codex, and reference guides you'll keep using.
For people who already work agentically and want to do it reliably, at team scale.
What you’ll learn
You already build agentic analyses for yourself. Learn to make them reliable enough to stand behind in front of a team — across models, with
In-depth understanding of open-source models and how they can be leveraged for agentic data analysis
Run frontier and open-source models through Claude Code, locally and in the cloud
Catch a quietly-wrong analysis before it reaches a stakeholder, systematically
Apply the four eval types to any AI generated data analysis
Multi-model validation - bake models off against each other and judge their analytical job
Multi-layered context engineering with freshness tracking and confidence grading, so your AI works across long sessions
Your system stays accurate and gets better over time instead of going stale
Take home your Claude Code AI Analyst system compatible with open source models
Take home a Codex-compatible version in AGENTS.md format. Same system.
Learn directly from expert instructors
Shane Butler
Principal Data Scientist @ Ontra | Ex-Stripe, Nextdoor, PwC, Appfolio
Sravya Madipalli
Data Science Leader@Superhuman (Prev. Grammarly)| Ex-Microsoft, eBay, Nextdoor
Hai Guan
Head of Data @ Ontra | Ex-Nextdoor, LinkedIn, Pinterest, Meta
Who this course is for
Product managers already using Claude Code for analysis. You've built skills and agents. Now build a system that validates its own output.
Data scientists who know how to work with Claude Code. Now automate the full workflow: context loading, analysis, validation, and narrative.
Analysts and engineers comfortable with skills and agents who want to go deeper in Claude Code.
What's included
Live sessions
Learn directly from your instructors in a real-time, interactive format.
Claude Code AI Analyst + Codex Repos(45+Skills and 25+Agents)
A working system. You get a configured CLAUDE.md that turns Claude Code into your analyst.
8 hours of live instruction (Saturday + Sunday, 4 hours each)
Saturday + Sunday, 4 hours each day. One instructor leads. Two troubleshoot and help you in real-time so nobody gets stuck. Every section is hands-on, you are building, not watching. Recordings of both days sent to all participants
Alumni community
Join a community of bootcamp alumni. Ask questions, share findings, get help troubleshooting, and see how others are using their AI analyst repos at their companies. Stay connected as the tools and techniques evolve.
Step-by-step guides for every section
10 reference guides covering best practices, Python scripts, and practitioner workflows
Lifetime access
Go back to course content and recordings whenever you need to.
Certificate of completion
Share your new skills with your employer or on LinkedIn.
Maven Guarantee
Your purchase is backed by the Maven Guarantee.
Course syllabus
7 live sessions • 31 lessons • 2 projects
Week 1
Jun
27
Bootcamp Day 1: Models I + Validation I
Ensure reliability on AI generated analysis
Swap the brain: open-source and multi-model in Claude Code
Validation I: Catch a Wrong Number Before You Ship It
Jun
28
Bootcamp Day 2: Validation II
Validation II: Score the System, Monitor It Over Time
Free resources
Schedule
Live sessions
8 hrs
Sat, Jun 27
2:00 PM—6:00 PM (UTC)
Sun, Jun 28
2:00 PM—6:00 PM (UTC)
Wed, Jul 1
3:00 PM—5:00 PM (UTC)
Frequently asked questions
Maven for Teams
Reimbursement
Get your company to pay
Everything L&D needs: email template, receipts, and certificate of completion.
Get reimbursedPrivate cohort
Run a cohort for your org
A dedicated cohort with a custom schedule and curriculum, tailored to your team.
Book a private cohort