Watch, Read, Listen
-
Everyone building AI coding workflows right now seems to converge on the same pattern: And to be clear: these help. But after watching one of my coding agents decide to make a direct change to AWS infrastructure through the AWS API — despite explicit instructions that infrastructure changes must go through Terraform — I started
-
this post formatted by AI for clarity, typos and readbility For the last fifteen years, software engineering has largely been a search for leverage through abstraction. Frameworks promised that if we accepted enough structure, conventions, lifecycle rules, and tooling complexity, we could dramatically reduce the amount of code humans had to write and maintain. And
-
One of the easiest mistakes for newer developers — especially in the age of vibe coding and AI-assisted app building — is storing uploaded images directly in the database. And honestly, it’s not just beginners. Even experienced engineers say things like: “We’re not going to have that many images.” “It’ll be simpler if everything is
-
/** AI used to format this article and clean up typos **/ The Prompt That Changes Things If you’ve got an API, try telling your AI this: “I want a Postman suite that gives me complete coverage of my API, that I can run locally but also runs as a GitHub Action when anyone makes
-
What Prompted This Post Two days ago, the founder of PocketOS shared a now-viral account of how an AI coding agent deleted his company’s production database—and the backups along with it—in a single automated action. With no viable recovery path, the team was forced to restore from a three-month-old backup that had only existed for
-
Most product teams have a quiet problem they don’t talk about enough: Everyone is working from a slightly different version of the product. Sales is pitching what they think exists.Product is planning what should exist.Engineering is building what was specified—weeks or months ago. And then AI shows up, and suddenly that ambiguity isn’t just inefficient—it’s
-
TL;DR Put this prompt into your AI Assistant: “I want to increase observability in my legacy project, and make it a critical component of all future development using speckit.” If you’re managing an engineering team, you already know this pattern: The issue isn’t a lack of data. It’s a lack of structured insight. Combining AI
-
I thought I had already learned this lesson. I had written it down once: do not put OPENAI_API_KEY in a global shell environment when using Codex over VS Code Remote SSH. ChatGPT billing and OpenAI API billing are separate systems. If Codex sees an API key, it may use API billing instead. That mistake cost
-
See an important follow up to this post first! I wanted a simple system: watch Gmail for messages that actually matter—especially job application replies—and send a WhatsApp alert when something actionable shows up. The first version worked. It was also unnecessarily complicated. What I really needed was something predictable, cheap to run, and easy to
-
I’ve been a ChatGPT Plus user for about two years. I’m a developer, and I use Codex constantly—it’s easily my preferred AI for coding. Most of my work happens over Remote SSH on a headless Ubuntu box I run called NATHAN. It sits quietly on my local subnet behind a firewall, doing real work. This