context-engineering-bootcamp

KO | EN

gitlite — search

#python #java #python3 #arduino #golang #machine-learning #rust #html #flask #javascript #seismology #nodejs

context-engineering-bootcamp

★ 14 Open GitHub ↗

Context Engineering for LLMs: structured notes on RAG, system prompts, MCP, Agents, and Memory Management — compiled from multiple expert sources.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# LLM Context Engineering Bootcamp

Comprehensive notes and visual resources on LLM context engineering, compiled from multiple sources. The primary resource is the [LLM Context Engineering Bootcamp](https://www.youtube.com/playlist?list=PLPTV0NXA_ZSj-E8A1DBvbWIYGxC46u-Hd) by [Vizuara AI Labs](https://context-engineering.vizuara.ai/) (taught by Dr. Sreedath Panat, PhD MIT), supplemented by additional tutorials, papers, and community resources.

## What Is Context Engineering?

Context engineering is the discipline of designing, assembling, and managing everything that flows into a large language model's context window to maximize output quality. It goes beyond single-turn prompt engineering to encompass system prompts, memory, retrieval, tool definitions, conversation history, and multi-agent coordination.

## Bootcamp Contents

> Notes marked with :white_check_mark: are complete. Notes marked with :construction: are in progress.

| # | Session | Topics | Status |
|---|---------|--------|--------|
| 1 | [**Introduction to LLM Context Engineering**](day_01/) | Prompts vs. context engineering, LLM OS analogy (Karpathy), six elements of context, context rot, lost-in-the-middle effect | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_01_notes.pdf) |
| 2 | [**System Prompts & CLAUDE.md**](day_02/) | System prompts at the "right altitude", CLAUDE.md / AGENTS.md / skill.md, iterative construction, few-shot example selection | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_02_notes.pdf) |
| 3 | [**RAG from Scratch**](day_03/) | WSCI framework, RAG pipeline end-to-end, chunking strategies, hybrid retrieval (dense + BM25 + RRF), cross-encoder reranking | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_03_notes.pdf) |
| 4 | [**Tools, MCP & Agents**](day_04/) | Tool schema design, Model Context Protocol (MCP), JIT instructions, ReAct agent loop | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_04_notes.pdf) |
| 5 | [**Context Engineering Failure Modes**](day_05/) | Context bloat, context rot, context poisoning, context clash, context distraction, context confusion, context drift, diagnostic framework | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_05_notes.pdf) |
| 6 | [**Multi-Agent Context Management**](day_06/) | AGENTS.md, compression, isolation, sub-agent architecture, central brain pattern, fan-out/fan-in, sequential pipelines | :white_check_mark: [PDF](https://di37.github.io/context-engineering-bootcamp/day_06_notes.pdf) |
| 7 | **Memory Management** | Episodic / semantic / procedural memory, windowing & summarization strategies, memory decay, cross-session persistence | :construction: In progress |

Other contents will be updated in the table as we continue to progress.

## Repository Structure

```
context-engineering-bootcamp/
├── README.md
├── day_01/
│   ├── day_01_notes.typ          # Typst source
│   ├── day_01_notes.pdf          # Compiled PDF
│   ├── brand.typ                 # Shared branding template
│   ├── references.bib            # Bibliography
│   ├── illustrations/            # Slide captures and figures
│   └── mermaid/                  # Mermaid diagram sources + renders
├── day_02/
│   ├── day_02_notes.typ
│   ├── day_02_notes.pdf
│   ├── brand.typ
│   ├── references.bib
│   ├── config-files-comparison.* # Platform config file comparison
│   ├── illustrations/
│   └── mermaid/
├── day_03/
│   ├── day_03_notes.typ
│   ├── day_03_notes.pdf
│   ├── brand.typ
│   ├── references.bib
│   ├── mteb_leaderboard.*        # MTEB embedding model leaderboard data
│   ├── illustrations/
│   ├── mermaid/
│   └── drawings/                 # Hand-drawn diagrams (retrieval, RAG patterns)
├── day_04/
│   ├── day_04_notes.typ
│   ├── day_04_notes.pdf
│   ├── brand.typ
│   ├── references.bib
│   ├── illustrations/
│   ├── mermaid/
│   └── drawings/
├── day_05/
│   ├── day_05_notes.typ
│   ├── day_05_notes.pdf
│   ├── brand.typ
│   ├── references.bib
│   ├── illustrations/
│   └── drawings/
└── day_06/
    ├── day_06_notes.typ
    ├── day_06_notes.pdf
    ├── central_brain_pattern.*
    ├── brand.typ
    ├── references.bib
    ├── illustrations/
    ├── mermaid/
    └── drawings/
```

## Prerequisites

- Basic familiarity with Python
- An API key from Anthropic, OpenAI, or Google Gemini
- Conceptual understanding of LLMs and next-token prediction
- For Day 3: familiarity with vectors, cosine similarity, and a Hugging Face access token

## Notes Format

Each day's notes are available in two formats:

| Format | Description |
|--------|-------------|
| `.pdf` | Compiled document with all illustrations and diagrams embedded |
| `.typ` | [Typst](https://typst.app/) source — compile with `typst compile day_XX_notes.typ` |

## Acknowledgments

The primary source is the [LLM Context Engineering Bootcamp](https://www.youtube.com/playlist?list=PLPTV0NXA_ZSj-E8A1DBvbWIYGxC46u-Hd) by Vizuara AI Labs. Additional content is drawn from various tutorials, research papers, and community resources. These notes are personal study materials created while learning from these sources.

context-engineering-bootcamp

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README