Sudo Make Me Sandwich
Join me as I write about technology, life, and everything in between.
An in-depth introduction to skills, what they are, where they live, and how they optimize context window utilization compared to traditional state injection.
Recent Posts
View All→
The Architecture of Performance: Optimizing Spark with Parquet Delta and Iceberg
Master the intricacies of modern data lakehouse storage formats to eliminate bottlenecks and slash compute costs. We dive deep into Z-Ordering, hidden partitioning, and the mechanics of file skipping.
Part 8: The Production RAG Master Class — Building Systems that Scale
The grand finale of our RAG series. Bringing together advanced chunking, hybrid retrieval, and architectural patterns into a unified production-ready framework.
The Magnum Opus: Architecting the Sentient Web with Next.js, WebGPU, and Hybrid RAG
An exhaustive architectural treatise on building "Sudo Make Me Sandwich". We dissect every atom of the stack: from the mathematical foundations of vector search to the silicon-level optimizations of WebGPU execution.
Part 7: Mastering Agentic Workflows — Moving Beyond the Chat Box
A step-by-step guide to supercharging your development efficiency using AI Agent rules, custom skills, and automated workflows.
Navigating the Invisible Walls: Engineering Around Free Tier LLM Rate Limits
Learn how to architect robust systems that survive the restrictive throttles of free LLM APIs through distributed state management and advanced queuing strategies. This guide moves from basic HTTP 429 handling to complex token bucket implementations for production-grade resilience.
Scaling Intelligence: A Production Guide to Deploying MCP to Vercel
Transitioning Model Context Protocol servers from local development to production environments requires a robust serverless strategy. This guide explores how to bridge the gap between local stdio protocols and the web-based transport layers necessary for Vercel deployments.