A list of courses does not explain how learning fits together
The Gen AI Roadmap turns learning into a visual prerequisite graph. It has four paths with different finish lines: AI Explorer, AI Engineer, Model Builder and Creative Studio. Someone learning to evaluate AI in everyday work should not be forced through the same sequence as someone who wants to build a transformer from scratch.
The roadmap describes 94 quests across 17 stages. A quest includes a mission, a short reason for its place in the sequence, concrete steps, an observable definition of done and learning links or search queries. The intention is to make each card an action rather than another bookmark to forget.
Prerequisites create branching paths, not a single ladder
Quests are connected by dependencies. A quest becomes ready when its prerequisites are complete; otherwise it stays visible and names what is blocking it. This keeps the map legible while preserving the learner’s ability to inspect what lies ahead.
The time-budget planner adds another practical constraint. A learner can choose a half-hour, evening, day or weekend and receive a run of quests that fit. The planner respects prerequisites as each step unlocks the next, instead of recommending a capstone that cannot yet be started.
Local progress is portable and privacy-conscious
The project is a static React, TypeScript and Vite site. It has no backend, account or database. Progress is stored in localStorage and can be exported as JSON, so a learner can keep a copy without sending a learning history to a service.
The app also supports direct URLs to a path and quest, print-specific output, and a phone layout that changes the roadmap graph into a stage list. The source README reports 58 tests for core roadmap behavior, alongside its build pipeline.
The roadmap is a starting map, not a universal curriculum
AI tooling changes quickly, so static links and dated recommendations need maintenance. The roadmap addresses that by using live search queries where a resource URL could not be checked, but learners should still evaluate the material for their goals and current versions.
The useful principle is that a roadmap should expose its assumptions. A visible prerequisite, a concrete completion check and an exportable progress file are more honest than a single percentage that pretends every learner is at the same starting point.
INSPECT THE IMPLEMENTATION
Read the code and project notes
The details above describe the public repository and project status as reviewed for this portfolio. Follow the source to inspect the implementation, setup instructions and current limitations.
Roadmap data and implementation
