Justin Ranton

Principal Design Architect | builds AI systems orgs actually adopt

Stop Pasting Specs Into Chat: Build Agent-Ready Repos

Facebook

AliveCor

SmarterDx

Allscripts

See a real spec fail live, then learn the three things an AI needs that a shared doc never provides.

How to spot why AI keeps losing your product context

Bring a real spec, PRD, or design doc. You'll migrate it in the hour. No git experience needed.

How to move a spec into a Git repo without engineering help

Leave with a Git repo, an AI-executable spec, and a link your team and your agents both work from.

How to build a repo your AI and your team both use

Build AI Agent Skills that Actually Work - 3 Hours Hands-on Workshop

Product Leader and Super IC at Google

Google

Disney+ Hotstar

Zynga

Anshumani Ruddra

Product Super IC at Google with 22+ years crafting experiences and blockbuster products

Cohort 07

Cohort 08

Most product builders use AI agents the same way they used search - typing queries and hoping for the best. That's leaving most of the value on the table.The teams pulling ahead aren't using better prompts. They're teaching their agents. Agent Skills are reusable, testable instruction sets that tell AI agents exactly how to handle specific workflows: your deployment process, your code review standards, your team's edge cases. Write one skill well, and every agent on your team benefits from it, every time, automatically.The problem: nobody is teaching product builders how to create them. Most documentation is written for a narrow technical audience, and the tooling is fragmented across Claude Code, Codex, and Gemini.In this hands-on workshop, you'll go from zero to a working skill installed in your own agent: one that reflects your real workflows, not generic templates. You'll also learn how to evaluate and improve skills systematically, so they get better over time instead of drifting.If you build or are hoping to build products with AI agents: this is the highest-leverage skill you can develop right now.

🤖 Learn to build procedural Agent Skills, transforming your professional mindset from solo prompter to manager of elite AI agents

Shift to an AI Management Mindset

Build Custom, Reusable Agent Skills

Leverage the Open-Source Ecosystem

Evaluate and Iterate for 10x Output

Product Managers: who use Claude Code or Codex daily but rely on ad-hoc prompts - and want repeatable, team-shareable AI workflows.

Entrepreneurs &amp; Founders: building AI-native products who need their agents to encode business logic, not just generate code on demand.

Tech Executives &amp; Product Leaders: who want their teams to stop reinventing prompts and start shipping a shared skills library.

The Conductor's Playbook: the six moves you'll own by the end. The tools change every few months; these are what you keep.

18 AI Products · 50,000+ Students

Apple

Nathan Wang

Founder of AI-Nate, a one-person company run by 30+ AI agents | 18 products in 18 weeks | PhD, Apple & Meta

AI Superpower Cohort 3

AI Superpower Cohort 4

Is this you?• You have watched 50 AI demos and built nothing you can point to.• You are stuck in endless ChatGPT threads instead of putting agents to work while you sleep.• Your team still does by hand what agents should be doing, and you are expected to have the answer.• You want something of your own, but "raise money, hire a team" feels like the wrong decade.The problem is not AI, and it is not you. Nobody taught you to conduct it.One person directing an agent crew can now do what used to take a team. True inside a company, for a consulting practice, or for a product of your own. The bottleneck was never ambition. It was headcount, and agents just removed it.In 4 weeks you build one real AI agent product, genuinely yours, on your own domain. Not a demo. Take it to work, show clients, or seed your own company.The skill outlasts the tool. ClawBot, Claude Code, whatever ships next, all just instruments. What you keep is the Conductor’s Playbook: how to brief, staff, and run a crew of agents toward an outcome, across Product, System, and Business.Cohort 4 is capped at 10. 6 seats are open. Doors close Sep 20.🚨 <a target="_blank" rel="noopener noreferrer nofollow" href="https://maven.com/ai-nate/ai-superpower/4/join?promoCode=EARLYBIRD">Self-funding? Code EARLYBIRD takes 20% off through Sun Sep 6.</a>

Your AI agent crew builds. You conduct. At work, or on your own product.

Conduct, Don't Do: Build a Product People Actually Want

One Job Per Agent: Master AgentOS From First Principles

Brief Before Build: Run It as a One-Person Company (OPC)

Ship Ugly, Ship Daily: A Real Product on Your Own Domain

The Handoff Is You: Orchestrate a Crew That Hands Off Work

Give It a Heartbeat: Agents That Run on Their Own Loop

Professionals Who Lead AI at WorkPMs, engineers, and team leads: ship a real agent product and bring the playbook back to your team.

Domain Experts &amp; EducatorsCoaches, professors, consultants: turn what you know into an AI product built by a crew, no engineers required.

Founders &amp; Side-BuildersLaunch your own product and run it as a one-person company: agents do the work, you own the outcome.

Build and ship a real product with Lovable, Supabase, Claude Code, n8n, and Hermes, then evaluate it with TRACE, across 4 hands-on weeks.

Cohort 4 [closed]

Cohort 5 [live]

PhD in ML | Google AI Accelerator Alum

Amazon Web Services

OpenAI

NVIDIA

Dr. Aki Wijesundara

AI Founder | Educator | Google AI Accelerator Alum

Exited AI Founder | Founder, TAI Labs

Boston Consulting Group (BCG)

McKinsey & Company

Manu Jayawardana

AI Advisor | Founder, TAI Labs

Cohort 3 [complete]

Cohort 6

Stop waiting on engineering. Ship your own AI product in 6 weeks.In six weeks, you build and ship one real AI product, no code required. You do not write code. You learn to direct it, verify it, and ship it, the skill that outlasts every tool.<ul><li>A live full stack app, built from a brief, not a spec nobody reads</li><li>Real data and user login, wired up without writing it by hand</li><li>A feature you direct an AI coder to build for you</li><li>Automation and a working agent, connected to tools you already use</li><li>Evals with TRACE, so you can prove the thing is actually good</li></ul>You will work with Lovable, Supabase, Claude Code, n8n, and Hermes. Most students arrive with the same problem: the idea was there, but shipping it meant waiting on engineering. You leave with a product you actually shipped.A limited number of need-based scholarships are available for applicants facing real financial hardship. <a target="_blank" rel="noopener noreferrer nofollow" class="underline underline underline-offset-2 decoration-1 decoration-current/40 hover:decoration-current focus:decoration-current" href="https://theaiinternship.com/scholarship">See if you qualify.</a>

PRICE GOES UP TO $1,999 IN 24 HOURS. OFFERS CAN'T BE COMBINED.

Ship a working product from a brief, not a spec

Direct AI to build, without writing code

Automate the work, and operate a live agent

Evaluate an AI product like a leader, with TRACE

Carry one product from idea to shipped

Lead an AI build instead of waiting on one

The one thing you really need. If you have a product or a process in mind and the drive to ship it, you are ready to start.

Curiosity and something you want to build

This is a no-code track. You will direct AI tools to build, not write code by hand. Comfort using everyday software is all you need.

No coding background required

We use Lovable, Supabase, Claude Code, n8n, and Hermes. Most have free or low-cost tiers. The pre-check session gets your accounts working so no live time is lost.

The tools, set up before we start

The PM who wants to ship, not just spec. You are done writing requirements and waiting on a backlog. You want to build it yourself.

The founder who needs a product, now. No technical co-founder, no budget for devs yet. You want a live product in front of users in weeks.

The operator automating real work. You see the manual processes everywhere, and you want to automate them.

Your AI Writes Empty Specs. The Context Layer Is Why — and How You Fix It

AI context systems for BAs & POs

Lufthansa Systems

Backbase

Luxoft

UKRSIBBANK BNP Paribas Group

Illia Shapoval

AI-assisted requirements engineering | 15 yrs PM in Banking, Aviation, MedTech

Cohort 1

Cohort 2

This workshop is about the context layer — the part that decides whether AI output is usable or just fluent.Skip it, and you get specs that are polished on the surface and empty underneath: generic phrasing, invented fields, rules your product never had.Overcorrect — "just give it all the context" — and it's worse. Do you even have all the context in one place? Would you want every Confluence page, every PRD revision, every deprecated business rule shaping the answer?The usual fixes all carry a bill:<ul><li>RAG pipelines — powerful, but now you build and babysit infrastructure.</li><li>Mega-prompts — thorough, but every answer comes out diluted.</li><li>Docs dumps — comprehensive, but light on the specifics that matter.</li><li>Letting AI manage its own context — convenient, right up until it takes the wheel.</li></ul>This workshop gives you a Context Engineering system that costs the least to start and grows with your product.You leave with a working starter context base, your own agents.md, your SKILL.md files, the feature you bring specced end to end, and a system you shape as you go, not a framework you serve. Bring a task you'd start Monday.I spent months sharpening this. Now it saves me days.

Engineer a compact context system that turns a four-line brief into a spec draft your team trusts — on your product, in your template.

Map requirement qualities to context blocks

Build an addressable, indexed context base for your product

Wire the base into a drafting system with agents.md, skills, and templates

Run the context-enrichment flywheel so the system stays alive

You're a Business Analyst or Requirements Engineer who owns product documentation, and you want a system for AI-assisted requirements work

You're a Product Manager or a PO who needs AI as a specification sidekick, not a chatty assistant you re-brief from scratch on every task.

You've tried "shove all the context in" — and watched the output get more generic, not less. Ready for a tested framework that fixes it.

Agentic AI

Coding with AI

AI Workflows

Claude Code

OpenClaw

Vibe Coding

AI Evals

AI Transformation

RAG & Search

AI for PMs

AI for Engineers

AI for Designers

AI for Marketers

AI for Founders

Product Sense

Product Discovery

User Research

Prototyping

Growth

Analytics

Tech Foundations

Strategy

Influence

Leadership

Career Growth

Machine Learning

LLM Ops

Context Eng

Security

System Design

Figma

Design Systems

Visual Design

Design Strategy

Automation

Content Marketing

Demand Gen

Go-to-Market

Product Marketing

Positioning

Social Media

Brand

B2B Marketing

SEO & AEO

AI for Leaders

AI Governance

Communication

Management

People Operations

Exec Presence

Storytelling

Goal-setting

Personal Brand

Fundraising

Everyone

Operators

Data Scientists

Business Analysts

User Researchers

Customer Success

Project Managers

HR Professionals

Sales People

Lawyers

Finance

Investors

Real Estate

Educators

Creators

Product managers who drive strategy, development, and launch of products.

Engineers and engineering leaders who build, deploy, and scale software systems.

Applying AI tools and techniques to professional workflows — assign alongside more specific AI topics.

Building autonomous AI agent systems that take actions, use tools, and orchestrate multi-step workflows.

Automating professional workflows with AI tools — productivity, operations, and task automation.

Designing and optimizing the context window for LLM applications — prompt structure, memory, and information retrieval.

Stop Pasting Specs Into Chat: Build Agent-Ready Repos

What you'll learn

Why this topic matters

You'll learn from

Justin Ranton