Engineering notes
Practical, experience-based writing on AI, RAG, agents, cloud, security, and shipping software to production.
- Local AIJun 29, 2026 · 10 min
What it actually costs to do local AI development in 2026: GPUs, models, and the math against Claude and Codex
Local AI cost in 2026, priced out honestly: GPU options and VRAM limits, electricity, and the break-even math against Claude, Codex, and API spend.
Read → - HiringJun 26, 2026 · 8 min
The true cost of an employee now includes their token spend
In the AI era the cost of a hire is salary plus benefits plus token spend. Why a cheap, wasteful hire can cost more per unit of output than an expensive one.
Read → - Local LLMsJun 22, 2026 · 9 min
llama.cpp vs Ollama: which should you run local models with?
llama.cpp vs Ollama, decided: how the two relate, where each wins on ease, control, and Hugging Face model access, and why we lean llama.cpp for real work.
Read → - ViteJun 16, 2026 · 10 min
Next.js vs Vite vs Remix for a product in 2026
Next.js vs Vite vs Remix in 2026: the real tradeoff is coupling versus composability. When each is the right call, and why we lean Vite for products that value not being locked in.
Read → - PulumiJun 9, 2026 · 10 min
How easy is Pulumi to use, compared to other infrastructure-as-code tools?
Pulumi ease of use compared to Terraform, OpenTofu, CloudFormation, and AWS CDK: language model, learning curve, tooling, and the honest tradeoffs.
Read → - MCPMay 28, 2026 · 10 min
MCP explained for product teams: how it differs from plain tool calling, and how to get auth right
MCP (Model Context Protocol) for product teams: how it differs from plain tool calling, the architecture, and a security-forward take on auth.
Read → - EvalsMay 19, 2026 · 9 min
Evals for LLM apps: why you cannot ship a reliable AI feature without them
You cannot ship a reliable AI feature on vibes. How to build a small, honest eval set, run offline and online evals, and turn argument into measurement.
Read → - AI AgentsApr 28, 2026 · 9 min
Agent orchestration and delegation: when one AI agent should become many
Agent orchestration is powerful and overused. When to split one AI agent into an orchestrator and specialized workers, and when not to.
Read → - RAGApr 15, 2026 · 10 min
RAG chunking strategies that actually work
Most RAG problems are chunking problems. A deep dive on chunking strategies, when each fits, the size and overlap tradeoffs, and how to measure it.
Read → - PrototypingApr 7, 2026 · 8 min
Throwaway HTML: the fastest way to kill a bad product idea
Why disposable single-file HTML prototypes beat slide decks and design files for early product decisions, how AI speeds them up, and when to stop.
Read → - AWSMar 12, 2026 · 8 min
Setting up OIDC between GitHub Actions and AWS (stop storing long-lived keys)
GitHub Actions to AWS OIDC: replace long-lived access keys in CI with short-lived, tightly scoped credentials. The trust handshake, scoping, and pitfalls.
Read → - AIFeb 26, 2026 · 9 min
Why most AI pilots never ship
Most AI pilots produce a great demo and quietly die at the production gate. The reasons they stall, and how to scope a pilot that actually ships.
Read → - Agent-Driven DevelopmentFeb 18, 2026 · 7 min
Agent-driven development: how we actually use AI coding tools
Agent-driven development multiplies what a senior engineer can ship without giving up code quality. How we use Claude Code, Codex, and Cursor in practice, and how to adopt it on a team without creating technical debt.
Read → - AIJan 22, 2026 · 8 min
Build vs buy: custom AI or off-the-shelf?
Build vs buy for AI: when an off-the-shelf tool wins, when a custom build wins, and the total-cost questions teams forget to ask.
Read → - RAGDec 15, 2025 · 9 min
Shipping RAG to production: what actually matters
A RAG demo takes an afternoon. A RAG system real users trust is a different animal. The things that decide whether retrieval-augmented generation works in production: retrieval quality, evals, grounding, and guardrails.
Read → - AINov 18, 2025 · 8 min
How much does it cost to build an AI MVP?
What an AI MVP actually costs, and what drives the budget: scope, integration, and data, not the model. A scope-first framework to ship in weeks.
Read → - RAGOct 21, 2025 · 9 min
RAG vs fine-tuning: which one do you actually need?
RAG vs fine-tuning, decided: what each one actually solves, when to use which, and why most teams reach for fine-tuning when they should not.
Read → - AISep 23, 2025 · 8 min
How to add AI to an existing product without a rewrite
Most AI work is not greenfield. A practical guide to bolting AI onto a product you already have: where to start, the integration patterns that are safe, and the failure modes to avoid.
Read → - CryptoAug 19, 2025 · 7 min
Cutting blockchain RPC costs: lessons from a crypto infrastructure audit
How a crypto infrastructure audit found a bug driving tens of thousands per month in wasted blockchain RPC costs, how that waste hides, and what to check.
Read →