AgenticAiLabs/Ai-Engineering-Roadmap

Path to becoming a self-taught AI Engineer - a curated, open-source curriculum modeled after OSSU.

What it solves

This project removes the barrier of expensive degrees and paywalled content by providing a structured, free, and open-source path to becoming an AI engineer. It organizes high-quality, community-curated educational resources into a clear, step-by-step roadmap.

How it works

The curriculum is divided into modular tracks that guide you from foundational programming and mathematics to advanced topics like Large Language Models (LLMs), agentic AI, and Retrieval-Augmented Generation (RAG). You can follow the path sequentially or jump to specific sections based on your current knowledge, using recommended world-class courses and practical project-based learning.

Who it’s for

It is designed for self-taught learners who want to build a portfolio-ready skill set in AI engineering, ranging from beginners starting with their first lines of code to those looking to master modern AI advancements.

Highlights

  • Comprehensive curriculum covering everything from math and machine learning to modern agentic AI.
  • Modular structure that allows for self-paced learning and specialization.
  • Focus on both theoretical understanding and practical, project-based application.
  • Curated links to world-class resources from institutions like MIT, Stanford, and Fast.ai.
  • Community-driven approach that stays updated with current AI trends rather than static academic semesters.

Related

  • Project
    e2b-dev/awesome-ai-sdks

    An open‑source curated list of SDKs, frameworks, and tools for building, monitoring, debugging, and deploying AI agents, maintained by the e2b team.

  • Project
    mem9-ai/mem9

    mem9 provides a persistent, shared memory layer for AI agents, allowing them to retain context across different sessions, machines, and platforms through a unified API.

  • Project
    traceroot-ai/traceroot

    TraceRoot is an open‑source platform that records AI agent calls, flags bad behaviour with custom detectors, clusters issues into signals, and runs automated evals so a coding‑agent can suggest and verify fixes, closing a self‑improvement loop. It offers a CLI, Python/TS SDKs, and integrations with major agent frameworks and model providers, and can be used via hosted cloud or self‑hosted Docker Compose.

  • Project
    wanxingai/LightAgent

    A modular, lightweight Python framework for building secure, multi-agent AI applications with support for complex workflows, persistent memory, and human-in-the-loop approvals.