v9ai/ai-engineer-roadmap

AI Engineer Roadmap is a hands-on learning platform that takes software engineers through the full journey of becoming a production AI engineer. The problem it solves is that AI engineering skills are scattered across papers, tutorials, and tools with no structured path from fundamentals to shipping real systems. This platform organizes 108 lessons across 15 categories, covering topics like RAG (Retrieval-Augmented Generation, a technique for grounding AI answers in real documents), agents, evaluations, fine-tuning, and prompting, ordered so each lesson builds on what came before. The platform is unusually feature-rich for a learning tool. It includes semantic search so you can find lessons by meaning rather than exact keywords, AI-generated audio narration for each lesson, an interactive knowledge graph that visualizes how concepts connect to each other, and a RAG-powered AI tutor you can chat with to ask questions grounded in the lesson content. It also tracks your mastery of each topic using a statistical model called Bayesian Knowledge Tracing, which estimates how well you know each concept based on your interactions. Someone would use this if they already know how to code and want a structured, deeply technical path into AI engineering, not just surface-level introductions. The tech stack is Next.js 15 on the frontend, Neon PostgreSQL with pgvector for vector similarity search, a Rust backend running LangGraph AI pipelines, OpenAI and DeepSeek for language models, and Cloudflare R2 and D1 for audio storage and playback state.