High-Dimensional Data Storage Talks

Two talks on vector search: from fundamentals to production-scale architecture decisions.

📖 Talks

Talk	File	Duration	Audience
Inside pgvector	pgvector_presentation.md	~20 min	How PostgreSQL stores, indexes & manages high-dimensional data
Vector Search Fundamentals	vector_search_fundamentals.md	~20 min	Getting started with pgvector — embeddings, indexing, semantic search
Storing High-Dimensional Data at Scale	vector_storage_at_scale.md	~20 min	Senior/architect — RAM wall, quantization, filtered search, DiskANN, architecture trade-offs

🚀 Quick Start

# Setup
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env  # Edit with your PostgreSQL credentials

# Generate demo data (50k docs, 1024d embeddings, ~10-15 min)
python scripts/generate_demo_embeddings.py

# Run a talk
presenterm vector_search_fundamentals.md
presenterm vector_storage_at_scale.md

📁 Structure

├── images/                              # Presentation images & gifs
├── scripts/
│   ├── generate_demo_embeddings.py      # Main data generator (50k docs)
│   ├── embedding_intro.py              # Embedding basics demo
│   ├── quantization_demo.py            # BQ compression + recall demo
│   ├── ram_wall_calculator.py          # RAM cost calculator
│   └── ...                             # Additional demo scripts
├── vector_search_fundamentals.md        # Talk 1: Fundamentals
├── vector_storage_at_scale.md           # Talk 2: Scale & architecture
├── archive/
│   ├── pgvector_deep_dive_demo.md       # SQL walkthrough demo
│   ├── vector_storage_at_scale_detailed.md
│   └── DEMO_COMMANDS.md                 # SQL commands for live demos
└── .env.example                         # DB config template

🔧 Configuration

Create .env file:

DATABASE_URL=postgres://user:password@localhost:5432/dbname

🎥 Presenter

Talks use presenterm. Install with:

cargo install presenterm
# or
brew install presenterm