Welcome to a collection of hands-on, developer-friendly tutorials covering recommender systems, geospatial data science, retrieval-augmented generation (RAG), and PostgreSQL full-text search. These guides focus on practical implementations using open-source tools and modern development practices.
Recommender Systems
From research prototype to production-scale recommendation engines
Build sophisticated recommender systems that handle millions of users and serve 15+ algorithms through unified APIs. This comprehensive series covers everything from graph database design to neural collaborative filtering.
Complete Journey:
- Architecture & Infrastructure - Multi-modal system design, graph schemas, and data pipelines
- Algorithm Deep-Dives - Content-based filtering, collaborative filtering, FastRP embeddings, matrix factorisation, and deep learning
- Production Engineering - Evaluation metrics, API design, and operational excellence
Featured Articles:
- Multi-Modal Architecture - Unified API serving 15+ algorithms with microservices design
- Graph-Native Collaborative Filtering - Sub-100ms queries across millions of relationships
- Universal Embeddings - Cross-type recommendations in unified vector spaces
- Neural Collaborative Filtering - When neural networks learn what linear algebra cannot
Perfect for: ML engineers, data scientists, and backend developers building recommendation engines that scale beyond academic datasets
Geospatial Data Science
PostGIS, Docker, GDAL, and spatial analysis workflows
Build robust geospatial data pipelines and perform advanced spatial analysis using industry-standard open-source tools.
Key Topics:
- Docker-based Geospatial Stack - Set up PostGIS, GDAL, and Jupyter in minutes
- DBSCAN Clustering - Density-based spatial clustering for irregular shapes
- PostGIS DBSCAN - Run clustering directly in SQL without Python
- OpenStreetMap & Overpass API - Extract custom geographic data with precision
- Data Conversion with ogr2ogr - Convert between geospatial formats and import to PostGIS
Perfect for: GIS analysts, data scientists, and developers working with location-based applications
RAG on a Web Domain
Chat with entire websites using open-source AI tools
Build a full-stack RAG pipeline that crawls, embeds, and enables conversational interactions with any website’s content.
Key Components:
- Quick Start Guide - Complete RAG pipeline overview
- Crawl4AI Implementation - Domain-aware web crawling and content extraction
- N8N & Supabase Setup - Workflow automation and vector storage
- Self-hosted LLM Deployment - Run your own models on DigitalOcean
Perfect for: Developers building AI-powered knowledge bases, chatbots, or content discovery systems
PostgreSQL Full Text Search
Powerful search without external dependencies
Master PostgreSQL’s built-in full-text search capabilities to implement sophisticated search functionality directly in your database.
What You’ll Learn:
- FTS Fundamentals - Complete guide to PostgreSQL search features
- Hands-on Tutorial - Practical examples with real-world datasets
- Advanced indexing strategies with GIN indexes
- Weighted search across multiple fields
- Result ranking with
ts_rankandts_rank_cd - Performance optimization techniques
Perfect for: Backend developers, database architects, and teams wanting to avoid external search infrastructure
Why These Tutorials?
Developer-First Approach: Every tutorial includes working code, Docker configurations, and real-world examples you can run immediately.
Open-Source Focus: No vendor lock-in. All tutorials use free, open-source tools that you can deploy anywhere.
Production-Ready: Techniques and patterns that scale from prototypes to production systems.
Modern Tooling: Docker, Git workflows, and cloud deployment strategies integrated throughout.
Getting Started
Each tutorial series is self-contained with its own setup instructions. Choose based on your current project needs:
- Need spatial analysis? → Start with Geospatial Stack
- Building AI applications? → Jump to RAG Quickstart
- Want better search? → Begin with FTS Tutorial
All code examples, Docker configurations, and sample datasets are available in the linked GitHub repositories.