SB / EDGE

LOG Writing — field reports

Notes from production.

What held, what broke, and what was cheaper the second time. Written after shipping, not before.

MAY 17, 2024

Designing AI-Powered Web Systems That Scale

Where the model belongs in the architecture, and where it very much does not.

ARCHITECTURE · 8 MIN
APR 26, 2024

The Evolving Role of the Frontend Architect

The job stopped being about components years ago. What it is about now.

FRONTEND · 7 MIN
APR 05, 2024

Integrating LLMs in Production Web Applications

Latency, cost, failure modes. The parts the demo never shows you.

AI INTEGRATION · 9 MIN
MAR 02, 2024

Practical AI: Tools, Not Magic

A model is a function with a latency budget. Treat it like one.

AI ENGINEERING · 6 MIN

EXT Published elsewhere

Local-first AI, in public.

FREECODECAMP

How to Handle Small Context Window Limits in RAG Systems

READ ↗
MEDIUM

The 3-Tier Memory System That Cut My AI Coding Context by 96%

READ ↗
DEV.TO

LLM as Router: Intent Classification for a Local Telegram Email Agent

READ ↗
DEV.TO

How Search and Ask Work: Local Hybrid RAG with ChromaDB + SQLite FTS5

READ ↗
DEV.TO

From Inbox to Character: Building a Private Local AI Email Agent

READ ↗