One canonical learning path through the full AI-SDLC — from foundations and core practices to building real products with AI at every phase. Start with the overview, then follow the tracks in order.
New to the series? Take the Overview to see how all 20 courses connect — then follow the five tracks below in order.
Every course builds a real component of the UCC Lien Risk platform. Work through five tracks in order — foundations, core practices, build & automate, and the leaders’ track. Each card shows a difficulty badge, and some offer a track toggle.
Master the domain, prompting, and your first AI coding tools — these make every later course far faster. No cloud accounts or prior code experience needed to begin.
The practices that make AI-assisted delivery reliable — spec-driven and context-driven development, AI project management, and measuring and governing adoption.
Build a real layer of the platform in your stack, then automate the whole loop. Data, API, and frontend courses each include a Cloud and a Local track; MCP servers and the AI Agent capstone tie it all together.
Specialized companions to the build courses — apply OpenSpec spec-driven development to one specific stack.
The leadership view — the business case, risk, and governance for adopting AI across your engineering organization.
Every course is anchored in a real-world use case: processing US Business UCC Filings to surface secured-transaction lien risk for commercial credit decisions. Here's the domain you'll master.
The Uniform Commercial Code Article 9 governs secured transactions in the US. When a lender takes collateral, they file a UCC-1 Financing Statement with the secretary of state — creating a public lien record that signals credit risk to other creditors.
Our platform ingests filings across 5 states, standardizes debtor names, links entities, and computes a composite Lien Risk Score (0–100) used in commercial underwriting.
A financing statement is seriously misleading if it fails the “standard search logic” test. Our pipeline must surface filings even when debtor names are misspelled, abbreviated, or use trade names — a core NLP challenge driving the entire data engineering course.
Whatever course you are in, the work maps to this cycle — plan, design, build, test, deploy, and govern, with AI at every step.