> ## Documentation Index
> Fetch the complete documentation index at: https://ai.tharung.in/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Engineering Roadmap

> Complete roadmap to become an AI Engineer

# AI Engineering Roadmap

Master:

* Python
* Machine Learning
* Deep Learning
* LLM Engineering
* RAG
* AI Agents
* MLOps
* System Design

<Card title="Start Learning" icon="rocket" href="#phase-0" horizontal>
  Follow the complete step-by-step AI Engineering roadmap.
</Card>

## Core Learning Phases

<Columns cols={2}>
  <Card title="Phase 0 — Orientation" icon="brain" href="#phase-0">
    Understand AI Engineering, market demand, and roadmap planning.
  </Card>

  <Card title="Phase 1 — Python" icon="code" href="#phase-1">
    Learn Python, async programming, NumPy, Pandas, and OOP.
  </Card>

  <Card title="Phase 2 — Mathematics" icon="square-root-variable" href="#phase-2">
    Linear algebra, calculus, probability, optimization.
  </Card>

  <Card title="Phase 3 — Machine Learning" icon="chart-line" href="#phase-3">
    Regression, trees, clustering, evaluation metrics.
  </Card>

  <Card title="Phase 4 — Deep Learning" icon="microchip" href="#phase-4">
    Neural networks, PyTorch, CNNs, transformers.
  </Card>

  <Card title="Phase 5 — NLP & Transformers" icon="language" href="#phase-5">
    Attention, embeddings, GPT, BERT, tokenization.
  </Card>

  <Card title="Phase 6 — LLM Engineering" icon="robot" href="#phase-6">
    Prompt engineering, APIs, streaming, routing.
  </Card>

  <Card title="Phase 7 — Multi-LLM Systems" icon="route" href="#phase-7">
    Build intelligent routing and orchestration systems.
  </Card>

  <Card title="Phase 8 — RAG" icon="database" href="#phase-8">
    Vector databases, embeddings, retrieval systems.
  </Card>

  <Card title="Phase 9 — AI Agents" icon="wand-magic-sparkles" href="#phase-9">
    Autonomous AI agents with tools and memory.
  </Card>

  <Card title="Phase 10 — Fine-tuning" icon="sliders" href="#phase-10">
    LoRA, QLoRA, RLHF, DPO pipelines.
  </Card>

  <Card title="Phase 11 — Generative AI" icon="image" href="#phase-11">
    Diffusion models, multimodal AI, voice AI.
  </Card>

  <Card title="Phase 12 — MLOps & LLMOps" icon="server" href="#phase-12">
    Deployment, observability, Kubernetes, monitoring.
  </Card>

  <Card title="Phase 13 — AI System Design" icon="diagram-project" href="#phase-13">
    Design scalable AI systems and architectures.
  </Card>

  <Card title="Phase 14 — SQL + pgvector" icon="table" href="#phase-14">
    Build vector search systems using Postgres.
  </Card>

  <Card title="Phase 15 — Quantization" icon="bolt" href="#phase-15">
    Optimize models for fast inference.
  </Card>

  <Card title="Phase 16 — Reinforcement Learning" icon="gamepad" href="#phase-16">
    PPO, RLHF, reward models, DPO.
  </Card>

  <Card title="Phase 17 — AI Safety" icon="shield" href="#phase-17">
    Ethics, governance, privacy, AI security.
  </Card>
</Columns>

***

## Featured Projects

<Columns cols={2}>
  <Card title="Multi-LLM Platform" icon="brain">
    Build production-grade AI orchestration systems with routing and fallback.
  </Card>

  <Card title="Enterprise RAG" icon="database">
    Multi-tenant RAG with reranking, HyDE, and vector databases.
  </Card>

  <Card title="AI Agent Platform" icon="robot">
    Autonomous research and coding agents with tool usage.
  </Card>

  <Card title="Production AI Infrastructure" icon="server">
    Kubernetes + monitoring + observability for AI systems.
  </Card>
</Columns>

***

## Learning Stack

<CardGroup cols={3}>
  <Card title="Languages" icon="code">
    Python, SQL, TypeScript
  </Card>

  <Card title="Frameworks" icon="boxes-stacked">
    PyTorch, FastAPI, LangChain
  </Card>

  <Card title="Infrastructure" icon="server">
    Docker, Kubernetes, Redis
  </Card>

  <Card title="Databases" icon="database">
    Postgres, Pinecone, Qdrant
  </Card>

  <Card title="AI APIs" icon="plug">
    OpenAI, Claude, Gemini
  </Card>

  <Card title="Deployment" icon="rocket">
    Vercel, AWS, GCP
  </Card>
</CardGroup>

***

## Final Capstone

<Card title="Production AI Platform" icon="stars">
  Combine Multi-LLM routing, RAG, Agents, Fine-tuning, Monitoring, and
  Kubernetes deployment into a single enterprise-grade platform.
</Card>