The archive / 2026
Essays.
Engineering practice, second-brain research, book reviews, and local models. Writing on this site, newest first. Earlier writing: Medium ↗ LinkedIn articles ↗
An Interactive Singapore Map, Built in One Conversation
A self-contained interactive map of Singapore - official planning boundaries, the full MRT network with an offline journey planner, statistics choropleths, georeferenced historical maps and walking tours - assembled from open data into one HTML file with an AI agent, no API keys. What the build produced, how it ran, and the token economics.
Teaching Every Harness the Same Skills
Three weeks turning one borrowed engineering method into a permanent capability: eighteen skills across every harness on two machines, one installer that caught its own bugs, a parallel-worker fleet lane proven end to end in twenty-four seconds, and the deletion of everything that had become a second copy of the truth.
The Devin Issue Pilot
Five days of turning Devin into an issue-driven teammate: seven PRs merged after midnight, a merge rule rewritten four times, a second Devin that reviews before any PR exists, and a weekly quota burned to 86% on a mode nobody checked. The timeline, the key learnings, and how it works now.
How Software Actually Works, in 95 Seconds
The eight layers of the How Software Actually Works class - terminal, files, apps, browser, deployment, git, agents and build day - in one short video, made in code with Remotion.
How Devin Fusion Ran a Five-Day Company-Brain Cutover
A 688M-token Devin Fusion session - Claude Fable 5.1 leading, SWE-2 as sidekick - cut over our company brain in five days. Why the job was so token-heavy, and why the saving over a single frontier model is mostly input, not output. With a two-minute narrated video.
The Last Project You Build Alone
Why a second brain should be custom to its owner, independent of any one AI harness, and able to give every future agent durable context.
Two Brains, One Operator
Why personal and corporate agent memory must split, how each half is governed - server-enforced write fences for the company brain, a permissive universal adapter for the personal one - and how Qwen 3.8 Max and Qwen 3.8 Flash Next on two DGX Sparks planned the architecture and audited 1,837 pages of company knowledge.
From Company Knowledge to Verified Changes
How Success Brain, shared abstractions, pattern linting, CodeGraph, and built-in verifiers work together to guide agentic engineering.
Why a Knowledge Management Layer Still Matters
A practical look at where no KM fails, where connectors and search help, and where a structured KM layer materially improves consistency, reasoning quality, traceability, and permissioning.
No matches. Try another term.
In the pipeline
-
Reviews
Book reviews - Victor Dibia, knowledge graph books, AI and systems titles. Roadmap on the Reviews track page.
-
Local Models
Local model and agent harness reviews, including Qwen-on-local-hardware evaluations. Roadmap on the Local Models track page.
-
Papers
Knowledge graph research - what a graph adds over plain retrieval. Listed under Second Brain and in this archive once published.