One Database, One Repo: Why Multi-Repo Hurts AI Coding
A multi-repo split I made for scalability reasons turned into duplicated code and handover files. My rule now: one database, one repo, period.
Stop Prompting in English If It's Not Your Language
Prompting an LLM in English when it's not your native language adds a translation layer between your thinking and the model — switch to your own language and that layer disappears.
Sometimes You Need to Let Them Go (And Sometimes You Need to Let Yourself Go)
Personal reflections on letting people go, being let go, and knowing when to leave a company yourself, drawn from the author's PM career.
What If AI Sat in the Middle of Every Conversation?
A WhatsApp voice message and an old failed startup called Alfred got me thinking: what if AI sat between every sender and every receiver, translating each message into the style the other person actually wants?
If You Want to Ride the Horse, You Need to Speak Its Language: A Non-Engineer's Survival Guide to Git and GitHub
A plain-language glossary and workflow guide to Git and GitHub for non-developers using AI coding tools.
GSD (Get Shit Done): The Framework That Changed How I Build Serious Software With AI
A deep dive into the GSD plugin for structuring multi-phase software projects with LLMs, from planning to autonomous execution and testing.
I Vibe Code Every Day Without an Engineering Team. Here Are the 3 Modes I Switch Between
Three modes of AI coding—auto-edit, planning, and GSD—and how matching the mode to the task drives better results without an engineering team.
Context Is Everything: Keeping AI Coding Agents From Losing the Plot
Why saturating a coding agent's context window quietly wrecks its output, and the docs/backlog/handover system I use to give agents real memory across a multi-repo project.
LLMs Don't Hallucinate — You Just Don't Ask Them to Show Their Work
The reframe that changed how I use AI for anything serious: it's not that the LLM hallucinates, it's that you never asked it to expand its context and prove its work. Here's why it happens — limited context, embeddings picking the closest match instead of the exact truth — and the two-step trick that kills most of it: expand the context, then demand proof for every sentence.
Permission Modes for AI Coding: Bypass, Auto-Edit, and Planning
Bypass, auto-edit, and planning are three different gears in Claude Code, not one setting — and knowing when to shift into planning is what saved a deploy bug I couldn't fix any other way.
Stop Calling It "Transparency"—Here's What Companies Should Actually Be
A critique of companies that claim transparency as a core value, arguing for being imperfect, candid, and human-first instead, with fact-based feedback.
There's a Better Way to Handle Layoffs (A Message to Leaders)
A plea to leaders to handle layoffs with honesty, real reasons, time, and help, and advice for those laid off to treat their career like a solo company.