AI 14 min read
My Agentic Web Development Setup Is Basically a `while True` Loop
A multi-model web development pipeline — research, design, implementation, review — eventually orchestrated by a boring Python while True loop.
36 articles
All writing
Engineering notes, long-form essays, experiments, and technical arguments—newest first.
Series
2 multi-part collections
AI 14 min read
The Dataset Is the Input
How prior-data fitted networks learn from synthetic worlds and make predictions on new datasets without task-specific weight updates.
AI 9 min read
Before You Build an AI Harness, Become the Harness
Before building software around an LLM, manually provide the tools and context it would have. The interventions reveal whether the workflow deserves a harness.
Linux 6 min read
I Left Arch + Niri for Pop!_OS COSMIC — and I’m Not Going Back (Yet)
After years on Arch with Niri, I switched to Pop!_OS COSMIC for .deb packages. Five days in, I still miss scrolling windows — not packaging.
Linux 9 min read
Your Desktop Needs `systemd --user`
Dotfiles and autostart launch processes but do not supervise them. systemd --user adds lifecycle management, restarts, timers, and logs to the Linux desktop.
Startups 9 min read
Cuegence Postmortem — Part III: Great Engineering Problems Don’t Make Great Businesses
Cuegence never failed, because we never built it. Part 3: the budget line, the replacement question, and why great engineering problems aren't great businesses.
Infrastructure 15 min read
I Moved Development to a Remote VM. Coding Agents Made Me Stay.
I moved my dev environment to a remote VM to stop rebuilding it across machines. Coding agents and tmux turned the experiment into how I work by default.
Startups 10 min read
Cuegence Postmortem — Part II: Six Schools and an Uncomfortable Answer
Three pivots, six calls with the people who actually sign purchase orders, one question: is this a budget line? Part 2 of the Cuegence postmortem.
Startups 7 min read
Cuegence Postmortem — Part I: I Fell in Love With the Engineering Problem
Curriculum graphs, evidence provenance, memory decay — a genuinely beautiful engineering problem. Part 1: how that beauty replaced the question of who would pay.
AI 12 min read
Carbon Chooses, Silicon Scales
AI may surpass humans at many kinds of thinking without replacing humanity. The real question is whether that intelligence stays accessible to everyone.
AI 16 min read
AI Feels Faster. The Data Is Less Certain.
AI coding tools feel transformative, but the productivity evidence is mixed. Part 2: perceived speed vs measured outcomes, and why review is the bottleneck.
AI 16 min read
AI-Native Software Engineering Isn’t About AI
AI-native engineering isn't about how much AI you use. It's a development model where AI is a primary participant in building software, not an assistant.
AI 18 min read
The Future of Software Engineering Is an Operating Model
AI-native engineering isn't about removing humans. Part 5: how roles, teams, economics and accountability shift when implementation stops being the main work.
AI 19 min read
The Model Is Only One Part of the System
A useful AI workflow takes more than a better prompt. Part 3: specifications, context engineering, agent workflows, and the harness that enforces boundaries.
AI 18 min read
When Agents Scale, Coordination Becomes the Product
One agent can misunderstand a requirement; several misunderstand it differently. Part 4: coordination, multi-agent orchestration and verification at scale.
Linux 12 min read
New to Tiling Window Managers? Start With Omarchy
Omarchy removes the dotfile rabbit hole and lets beginners experience a complete Hyprland workflow before deciding what they want to customise.
AI 16 min read
The Model Is Not the Development Environment
I spent 12 million input tokens on a small monorepo change. The problem was not the model. It was the development environment around it.
AI 11 min read
AI Speeds Up Coding, Not Engineering
An opinionated workflow for building good-enough software with AI coding agents without surrendering architecture, judgment, or security.
AI 18 min read
I Tried Too Many Coding Agents. Here Is What I Would Actually Pay For
A subjective comparison of Claude Code, Codex, Cursor, OpenCode, and the models inside them, based on ordinary application work rather than benchmarks.
Software Engineering 11 min read
I Couldn't Become a 10x Developer
Chasing 10x across six projects, several languages and a pile of AI agents — and the quiet realization that 1.5x is honest work.
AI 12 min read
Your AI Agent Should Be a Folder
After testing agent frameworks, I found the best personal AI agent isn't a platform — it's a well-organized folder of Markdown, scripts, logs and state.
Software Engineering 13 min read
We Abused POST for Years. HTTP QUERY Is the Apology.
Complex read-only queries have hidden inside POST /search for years. The HTTP QUERY method is safe, idempotent, and built for body-heavy reads.
Infrastructure 7 min read
Why Don't We Build Data Centers in Antarctica?
Antarctica is cold, so why no data centers? Because data centers are factories, and factories need power, fiber, roads, spare parts, workers and permits.
Infrastructure 9 min read
Why Don't We Build Data Centers in Space?
Free solar, no land disputes, no cooling towers. But space removes almost everything a data center needs: easy cooling, maintenance, upgrades and repair.
AI 6 min read
I Automated an Entire Course Production Pipeline. Then I Decided Not to Use It.
I built an AI pipeline turning storyboards into assets, narration, slides and captions. It almost worked, and showed why courses are authored, not manufactured.
AI 6 min read
AI Doesn't Need to Become a Better Doctor. It Needs to Become a Better Historian.
AI spends 30 minutes building a medical history; a doctor gets 10. The gap isn't intelligence, it's information. Build AI historians, not AI doctors.
Software Engineering 6 min read
Stop Cloning the Same Repo Again and Again. Git Worktrees Exist.
Stop cloning the same repo for every context switch. Git worktrees give you several branches at once — no WIP commits, no broken dependencies, no clutter.
Infrastructure 10 min read
Why I Created an Ultra-Simple MySQL Backup Tool
Backups shouldn't need dashboards, agents or cloud tokens — just a boring script, cron, and a compressed dump on another VM. The thinking behind mysql-backup.
AI 9 min read
Human-in-the-Loop Is a Lie We Tell Ourselves Before Clicking "Approve All"
Human-in-the-loop sounds responsible, but tired humans rubber-stamping prompts isn't governance. Real answer: programmatic boundaries, human accountability.
Software Engineering 11 min read
Why Is Laravel So Underrated?
Laravel gets dismissed as outdated PHP, yet it still ships serious business applications faster than most modern stacks. Why it's still my backend default.
AI 6 min read
Learning in an AI-Driven World is a Paradox
AI changed the feedback loop between learning and building. I ship faster than ever, yet some days I know less than ever. What that paradox means for you.
Linux 8 min read
Why I Chose Niri + DMS—and Why I’m Staying for a While
After years of tiling window managers, Niri's scrollable tiling and DankMaterialShell finally fit how I actually think about projects and contexts.
AI 8 min read
I Tried Building My AI Agent With Existing Agent Frameworks. Then I Chose the Boring Deterministic Way.
After evaluating multiple AI agent frameworks, I found that deterministic automation with an LLM inside beats magic autonomy for real operational work.
AI 7 min read
Not Everyone Needs Frontier LLMs for Coding
Frontier models are brilliant, but most everyday coding runs fine on smaller ones if your workflow is clear. Choose tools for your use case, not the hype.
Software Engineering 6 min read
I Built a Tiny Task Tracker Because Jira Was Too Much for My Animation Team
When Jira, ClickUp and OpenProject overwhelmed a non-technical creative team, I built a small app focused on backlog pull, visibility and simplicity.
AI 9 min read
Why Use LangGraph for Your First Agentic AI Project
LangGraph forces you to learn state, tools, routing and human approval — the agentic AI fundamentals that survive framework churn.
