With over two decades of experience in data systems architecture and recent years dedicated to AI engineering, I share insights from my experiences here. I hope you find these posts useful. Your comments and feedback are greatly appreciated. — Rares Vernica

LangGraph is MIT-Licensed, but Your Production Deployment Might Not Be

Posted in March 2026 by Rares Vernica

LangChain has quietly built the most capable stack for production AI agents. Between the graph-based orchestration of LangGraph and the new DeepAgents harness, the developer experience is elite. But if you’re planning to self-host, there is a “licensing boundary” you need to know about before you commit to the architecture.... [Read More]

Elevate Your LLM Applications: A Deep Dive into Advanced RAG Techniques 🚀

Building Production-Ready Systems with Advanced Retrieval Strategies

Posted in December 2025 by Rares Vernica

Retrieval Augmented Generation (RAG) has revolutionized how Large Language Models (LLMs) access and utilize external knowledge, moving beyond static training data to deliver more accurate, relevant, and up-to-date responses. While basic RAG setups are powerful, achieving peak performance and addressing complex real-world scenarios often requires a more sophisticated approach. This... [Read More]

Dual-Engine Serverless SQL Lakehouse

Powerful lakehouse with DuckDB, Neon PostgreSQL, and Google Cloud Storage 🌊🦆🐘

Posted in June 2025 by Rares Vernica

The concept of a data lakehouse is gaining serious traction, and for good reason! It combines the flexibility and cost-effectiveness of a data lake with the data management features and ACID transactions of a data warehouse. Let’s dive into building a practical lakehouse setup using some fantastic open-source and cloud-native... [Read More]

Machine Learning in SciDB

Leveraging the SciDB Stream plug-in

Posted in October 2017 by Rares Vernica

Popular data processing platforms offer users the ability to inject an external process into the data processing pipeline. The data flowing through the data pipeline is fed as input to the external process, while the output produced by the process is fed back into the pipeline. The external process runs... [Read More]