Siddharth B·2d agoAn Embedding Model Upgrade Is a Data MigrationHow to rebuild RAG indexes without silently corrupting retrievalA response icon3A response icon3
Siddharth B·Aug 22Microservices Are Not Defined by Where They RunWhen Tool Knowledge Gets Mistaken for Architecture
Siddharth B·Jul 11A Risk Signal Is Useless Until Someone Can Act on ItUsing Transactional Outbox and CQRS-Lite in an Early Warning System
Siddharth B·May 26Building a Production-Grade RAG Pipeline for BFSI Documents — Lessons from the TrenchesThis isn’t a tutorial. Enough of those exist. This is a post-mortem, stitched together from what teams actually encountered when the…
Siddharth B·May 16Why We Built Our Own Data Sanitization Layer Instead of Buying OneADR Series — Article 5 of 5
Siddharth B·May 15Why We Recommend Event Streaming Over Synchronous REST for Financial PipelinesADR Series — Article 4 of 5
Siddharth B·May 13Why We Recommend a Multi-LLM Architecture Over a Single ProviderADR Series — Article 2 of 5
Siddharth B·May 12Why I Recommend RAG Over Fine-Tuning for Credit AnalysisADR Series — Article 1 of 5A response icon1A response icon1
Siddharth B·Jan 5Stop Asking ‘Cloud or On-Premise?’ — Here’s the Question That Actually MattersI’ve been in enterprise technology for over two decades now, and if there’s one question that never seems to go away, it’s this: