What we think about AI, cloud, and building products.
Building a production MCP server that lets AI assistants query your business database safely. Full Python implementation with FastMCP, read-only SQL execution, schema inspection tools, and analytics prompts.
Building a production document processing agent step by step. PDF intake, classification, extraction, validation, and structured output using Claude, tool calling, and a retry loop.
A step-by-step build of a production CI/CD pipeline. GitHub Actions, ECR, ECS Fargate, canary deployments, automated rollback, and Slack alerts. Every decision explained.
How EDS changes content delivery in the Adobe ecosystem, where Universal Editor fits, and practical architectural decisions for enterprise implementations.
Agentic AI is more than a buzzword. Here is what it really means, how it differs from traditional AI, and where it delivers genuine business value.
Building a RAG pipeline that works in a demo is easy. Building one that works reliably in production is a different challenge entirely. Here is what we have learned.
Migrating from AEM on-premise or AMS to AEM Cloud Service is a significant undertaking. Here is a practical guide based on real migration projects.
Most retail software assumes you have reliable internet. In the real world, that assumption costs stores money every day.
AWS cost optimisation is not about penny-pinching. It is about right-sizing your architecture so you pay for what you use and nothing more.