Learning path map
Detailed content
📈 The Opportunity: AI + Consulting Now
Why this is the best time to get started, what AI consulting looks like in practice, and the thesis that guides everything: minimal input, maximum output.
Every company needs an AI strategy for 2026, but few know where to start. There's a gap between wanting to act and actually executing.
28,974 people pay for Wharton's AI strategy course alone. Demand is enormous—and those who deliver, not just talk, win the market.
Window of opportunity · execution vs. intention gap · first-mover advantage.
Diagnose where the company is, find where AI creates value, and deliver a concrete plan (roadmap, ROI, governance) that the client can execute.
This is exactly what the Factory automates. Understanding the service inside and out keeps you in control of what the machine produces.
Diagnosis · value opportunity · actionable deliverable.
You enter the company name and description and get 19 deliverables. The effort goes into building the machine, not into each deliverable.
It’s the principle that separates a freelancer from a product. Build once, deliver infinitely.
Leverage · productization · near-zero marginal cost.
Three paths for the same engine: use it in your company, sell it as a service, or productize it as a recurring offer.
Define how you’ll monetize. Jason Liu scaled an AI consulting business from $170/hr to $100k/month in one year by productizing it.
Internal use · project · retainer · productized offer.
Claude Code writes and runs the code. Your job is to direct it: describe what you want, review, and adjust.
Removes the barrier that holds most people back. Non-developers build the entire Factory by asking Claude Code.
Orchestration · vibe coding · review > write.
5 tracks = 5 levels. Each module ends with a practical Mission and gives you a reusable resource (skill, agent, prompt).
You always know where you are and what you’ve gained. Clear progression keeps you moving toward the capstone.
Levels · Mission with payoff · growing arsenal.
🔧 Anatomy of the Factory
The 19 deliverables, the 3-phase pipeline, and why the result looks like elite consulting. A map of what you'll build.
15 Markdown documents + 2 PPTX presentations + 2 Word documents. From the technical inventory to the closing SOW.
Knowing what comes out defines what you can sell. Each item is a deliverable clients pay a premium for.
Markdown · PPTX · DOCX · executive package.
Research (Perplexity) → Synthesis (Gemini, 15 prompts) → Generation (PPTX/DOCX/PNG). Each phase feeds the next.
It’s the course’s backbone. Each learning path explores one part of this pipeline in greater depth until you master the whole thing.
Pipeline · chained phases · separation of responsibilities.
Research gathers current company facts; Synthesis turns facts + frameworks into documents; Generation formats them for delivery.
Understanding the "why" behind each phase helps you avoid using AI as a black box—you know where to intervene and improve.
Current facts · guided synthesis · professional formatting.
The prompts were built using real playbooks from McKinsey, BCG, KPMG, OpenAI, and IBM, distilled into cheat sheets.
The output inherits the rigor of those consultancies. That's what makes the package look (and feel) like elite work.
Distilled knowledge · borrowed authority · perceived quality.
A tour of the files generated for a company: roadmap, ROI, AI policy, deck, and SOW.
Seeing the destination before the journey gives you a quality benchmark and concrete motivation.
Final deliverable · quality reference · big-picture view.
How the 5 tracks connect: knowledge → skills/agents → the factory → the client.
You move through each track knowing which part of the Factory you’re building at that moment.
Logical sequence · incremental build · capstone.
⚙️ Building your arsenal
Claude Code, accounts and keys (Perplexity + Gemini), clone and run the repo, the build-in-public mindset, and how to control costs. By the end, you’ll run the first package.
The development assistant that clones, installs, configures, and runs the Factory based on natural-language instructions.
It’s what makes everything possible without coding. Mastering how to make good requests is the course’s number one skill.
Natural language · install/run · iterate through conversation.
The API that researches the company online in real time. Generate a free key in Settings → API.
It’s what keeps the deliverables up to date—it doesn’t make things up; it finds recent facts about the company.
API key · live research · .env.
Gemini 2.5 Flash generates the documents: affordable and with a huge context window. Key created in Google AI Studio.
It’s the writing engine. Its low cost is what makes it possible to deliver an entire package for less than US$1.
Gemini 2.5 Flash · large context · low cost.
Download the ai-strategy-factory project, run setup, and launch the web interface at localhost:8888.
Ask Claude Code to "clone and run the Factory for me" — and watch it get up and running in minutes.
git clone · setup.py · webapp localhost.
Building in public: share the process, including mistakes (context limits, API, wrong model name).
Builds authority and attracts clients. Showing the actual process connects more than just the polished result.
Transparency · authority · proof of work.
Two research modes: quick (~US$0,05) and deep (~US$0,50). You choose based on your budget.
Controlling costs is what keeps margins high. Documentation is cheap in tokens; split up large requests.
Quick vs deep · token budget · margin.