11 articles
AI & Tooling
Practical coverage of AI tooling, agent workflows, local assistants, search, testing, and architecture decisions.
AI tooling is useful only when it improves a real engineering loop: understanding code, testing changes, searching internal knowledge, or making operational decisions less noisy.
Use this category when you are deciding whether an AI assistant, agent workflow, local model, vector database, or AIOps feature belongs in your stack.
Context Is the New Cache: How to Budget Tokens Like You Budget RAM
Don't treat large LLM context windows as an infinite scratchpad. Learn to apply classic cache design principles—like eviction, locality, and prefetching—to engineer smarter, more cost-effective, and performant AI applications.
Jun 11, 2026 · 10 min read
Build a Personal Developer AI Assistant with MCP, Continue, and Local Code Search
You do not need to build a chatbot from scratch to get a useful developer AI assistant. Use standard tools like Continue, MCP, ripgrep, and an LLM API to create an assistant that understands your repo and fits your workflow.
May 29, 2026 · 12 min read
The AI Architect Is Becoming Your Platform Copilot
LLMs are starting to reshape how you design, govern, and evolve internal developer platforms. Here is how to use them practically without turning your platform into an unreviewed pile of generated YAML.
May 27, 2026 · 11 min read
The 'Invisible IDE': Why Your Next Development Environment Won't Have a UI (and Why That's a Good Thing)
Say goodbye to bulky desktop IDEs and hello to a new era of developer environments that prioritize consistency, speed, and remote collaboration by shedding the graphical interface.
May 21, 2026 · 8 min read
Vector Databases Are Not Enough: Why Hybrid Search Is the Only Path to Production-Grade RAG
Early RAG systems relied heavily on vector databases, but for production-grade applications, pure semantic search proves brittle. Discover why hybrid search, combining embeddings with keyword and structured data, is the essential evolution for reliable RAG.
May 19, 2026 · 11 min read
The True Cost of Local LLMs: Beyond the GPU Price Tag
Running large language models locally offers exciting possibilities, but your GPU purchase is just the first expense in a much broader, and often hidden, cost landscape.
May 16, 2026 · 11 min read