AI-SHIPR

AI-native product operating system for product managers using Claude Code.

TL;DR

AI-SHIPR is a product manager's operating system — a folder-based system that gives Claude persistent memory of your product.

Instead of starting every session from zero, Claude reads your strategy, hypotheses, initiatives, and past decisions — so it can think with you, not just respond to prompts.

👉 It turns AI from a tool into a true product thinking partner.

The problem

Every time you open a new Claude session, you start from zero.

You re-explain:

• your product
• your strategy
• your current bets
• your team context

You get useful output — for that session.

Then it's gone.

AI helps with tasks.
But it doesn’t know your product.

What AI-SHIPR does

AI-SHIPR gives Claude persistent, structured context across your entire product workflow.

It organizes your work into a system Claude can read and update:

• Strategy → vision, bets, KPIs
• Hypotheses → structured, falsifiable
• Initiatives → linked to bets
• Proof → experiments with decision thresholds
• Relationships → users, stakeholders, PM profile
• Learning → accumulates across sessions

Every session builds on the last one.

See it in 10 seconds

Instead of:

“Let me explain my product again…”

You just run:

/today

And Claude:

• Reviews active initiatives
• Flags weak or broken hypotheses
• Suggests next experiments
• Uses your actual product context

Why this is different

Most people use:

• prompts
• templates
• static context files

AI-SHIPR is a harness, not a prompt:

1. It updates itself
Agents and workflows write back into the system after each session.

2. It enforces product thinking

• hypotheses must be falsifiable
• initiatives must link to bets
• experiments must have failure thresholds

3. It compounds over time
The more you use it, the smarter it gets.

Try it (2 minutes)

Open Claude and paste:

“What’s missing from this hypothesis?”

Then paste:
examples/Duolingo/H-Hypotheses/HYP-001.md

Now imagine doing this with your own product —
without re-explaining anything.

Who this is for

Product managers who:

• already use Claude or ChatGPT
• are tired of repeating context
• want AI to actually understand their product

What’s inside

AI-SHIPR/
├── S-Strategy/        → vision, bets, KPIs
├── H-Hypotheses/      → structured hypotheses
├── I-Initiatives/     → execution layer
├── P-Proof/           → experiments
├── R-Relationships/   → users, stakeholders
├── Learning.md        → accumulated knowledge
└── A-AI/              → agents, skills, workflows

See a full example

Check: examples/Duolingo/

A complete working system with:

• real product tensions
• competing bets
• experiments with thresholds
• linked artifacts across the system

Getting started

1. Clone the repo
2. Install Claude Code
3. Open Half-Sprint-Guide.md
4. Start with your Vision + Strategy

AI-SHIPR = Strategy · Hypotheses · Initiatives · Proof · Relationships

The full Agile loop — with memory.

For setup guides, Team collaboration setup, and additional resources: verve-pm.com/ai-shipr-resources

About

Yaniv Yaakubovich is a product management consultant based in Israel. He has led product at Google, PayPal, and early-stage startups across fintech, edtech, and SaaS. He now works with founders and product teams helping them build with clarity — clear strategy, hypothesis discipline, and AI that actually knows the product.

AI-SHIPR came out of his own consulting practice. He built it because he was tired of re-explaining his clients' products to Claude at the start of every session. He uses it daily. The workshop is how he installs it on other PMs' machines.

• Workshop: English verve-pm.com/ai-shipr | Hebrew verve-pm.com/ai-shipr-he
• Resources: verve-pm.com/ai-shipr-resources
• Consulting: verve-pm.com
• LinkedIn: linkedin.com/in/yanivy9h

AI-SHIPR stands for: Strategy · Hypotheses · Initiatives · Proof · Relationships
The full Agile loop, in one system.