20 articles and counting
Thirteen years as a PM taught me there are two kinds of disagreement. The kind about the product, and everything else that slips downstream and warps it. AI removed the second kind. What is left is pure accountability, and that is the joy.
If your website lives on a templated platform and you have ever wished you could change more, or take your content with you, this is the idea worth knowing. A plain-language look at headless CMS, who it is for, the trade-offs, and the main tools, written while I am still learning it myself.
Susan Street Fine Art represents decades of beautiful work and real collector relationships. The redesign was about one thing: letting the art, the artists, and the projects take center stage, with a fast, modern experience to match.
Program Edge grew from brand research to a live five-page site in two sessions — with a homepage that embeds the client build live, rather than showing a screenshot of it.
For Center X Equine Therapy, the invisible stack — Givebutter, Fillout, Airtable, Cloudflare, and a thoughtful HIPAA boundary — mattered just as much as the HTML.
The clients loved the demo — and most of the work that made that moment possible happened before I opened a code editor: competitor research, a discovery agenda, and the quiet trust patterns nonprofits need.
We're engineering bigger context windows, better RAG, smarter memory. Meanwhile the human — the one holding all the threads — is losing context faster than any model.
After a recent two-bug incident with a third-party data feed, I rewrote the rules I follow when integrating any external data source. Six items, two pieces of supporting infrastructure, and a clearer line between 'is the API up?' and 'is the data right?'
My market-risk agent told me the Strait of Hormuz was calm during a real, active crisis. Tracking down why turned into a same-day case study on the difference between an integration that runs and an integration you can trust.
I started with a familiar resident frustration: HOAs feel restrictive and hard to navigate. The useful product idea was not generic AI Q&A. It was California procedure, deadlines, checklists, and calm self-help.
A chart on my financial report showed one number while the header showed another. The fix was small, but the lesson was bigger: decision-support tools earn trust through boring, precise details.
I like resumes, but I love links. Here is how I use a live report, this blog, and an open knowledge base as a gentle way to show how I think and build—without a big marketing machine behind me.
My market report grew past a thousand lines, and one section carried most of the scroll. Here is how I made it kinder to read on a phone—without jumping straight to a big client-side framework.
When my financial report hit 100/100 CRITICAL risk and I had a real 401(k) question, I stopped treating the project as a learning exercise and started building for real stakes.
We published a blog post about 8 context engineering components. Then we audited our own workspace against the framework. Six of seven were strong. One was barely there.
The transformer paper told us how AI models think. Context engineering tells us what they see. These are two different layers of the same stack, and confusing them costs you.
Plan mode in Cursor isn't a solo pause — it's where human and AI align on direction before execution. Here's why that collaboration matters, and what I do when the UI and the chat disagree.
I named my AI agent Lumen, wrote 13 principles for how we work together, and built a financial risk engine in three days. Here's what actually happened.
I built this blog in a single session — from an empty folder to a live site on GitHub Pages. Here's the real build story, the gotchas, and why Astro was the right call.
I'm a project manager who built a financial risk engine, a memoir blog, and a knowledge base — all with AI agents. Here's what building taught me that frameworks never could.