AI Engineering¶
Technical deep-dives into AI system architecture and implementation patterns.
Building Hybrid Scalable Search Systems with Vector Databases¶
March 2025 • 8 min read
As AI-powered applications become more prevalent, the need for scalable search systems has never been greater. This article explores the architecture patterns and best practices for building search systems that can handle millions of users while maintaining sub-second response times.
Vector databases are specialized storage systems optimized for high-dimensional vector operations. They enable similarity search at scale, which is crucial for modern AI applications. The key to building scalable systems lies in understanding the trade-offs between accuracy, speed, and resource consumption.
Scaling Transaction Tagging: From Rules to ML at 10M+ Transactions¶
December 2024 • 10 min read
Building intelligent financial systems requires solving complex data challenges at scale. This article explores the architecture and implementation of an automated transaction tagging system that processes over 10 million transactions, transitioning from rule-based approaches to sophisticated ML pipelines.
The system implements a three-stage approach: lexical rule-based preprocessing, intelligent F1 tagging that learns from user behavior, and ML-based categorization using a fine-tuned DistilBERT model for semantic understanding.