AI / MLMarch 1, 20264 min read
The Evolution of AI Application Stack: 2020 to 2024
How RAG, LangChain, LangGraph, and GraphRAG built on each other to shape the modern LLM application stack.
Read ArticleBlog / Writing
Engineering Blog
Notes on voice AI, SaaS architecture, frontend performance, and full-stack engineering.
AI / MLJanuary 10, 20265 min read
RAG (2020): Retrieval-Augmented Generation
RAG retrieves relevant documents for LLMs, grounding generation in external knowledge instead of model memory alone.
Read ArticleAI / MLFebruary 5, 20265 min read
LangChain (2022): Building LLM Applications
LangChain is a framework to build LLM applications, released by Harrison Chase in October 2022.
Read ArticleAI / MLFebruary 20, 20266 min read
LangGraph (2024): Orchestrating Complex Agents
LangGraph is a framework to orchestrate complex agents built with LangChain, supporting cycles, state, and multi-agent workflows.
Read ArticleAI / MLMarch 5, 20266 min read
GraphRAG (2024): Advanced RAG with Knowledge Graphs
GraphRAG is advanced RAG using knowledge graphs, introduced by Microsoft Research in July 2024.
Read ArticleAI / MLMarch 12, 20266 min read
Applying the Stack: RAG + LangGraph in a Booking Platform
How RAG retrieval and LangGraph agent orchestration work together in a production AI booking platform.
Read ArticleAI / MLMarch 15, 20265 min read
Hybrid Search: BM25 + pgvector for RAG Retrieval
Combining Elasticsearch BM25 keyword search with pgvector semantic search for better RAG retrieval quality.
Read ArticleAI / MLMarch 18, 20265 min read
Choosing Between RAG and GraphRAG in Production
Practical guidance on when classic RAG is enough and when GraphRAG's knowledge graph approach adds value.
Read ArticleAI / MLMarch 22, 20265 min read
Migrating from LangChain Chains to LangGraph Agents
When and how to move from LangChain linear chains to LangGraph graph-based agent orchestration.
Read ArticleFull StackAugust 22, 20255 min read
Next.js Performance Patterns for Product Teams
Practical React and Next.js optimizations that improve real user experience without over-engineering the codebase.
Read Article