<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Blog on Amfelt</title><link>https://amfelt.com/blog/</link><description>Recent content in Blog on Amfelt</description><generator>Hugo</generator><language>en-us</language><copyright>© 2026 Amfelt Labs B.V.</copyright><atom:link href="https://amfelt.com/blog/index.xml" rel="self" type="application/rss+xml"/><item><title>Amfelt's first version is ready for closed testing</title><link>https://amfelt.com/blog/first-version-closed-testing/</link><pubDate>Sun, 01 Nov 2026 00:12:00 +0000</pubDate><guid>https://amfelt.com/blog/first-version-closed-testing/</guid><description>The first version of Amfelt is complete. We are now running demos and tests with a small group of invited teams.</description></item><item><title>The board wants AI. The data is still on the ship.</title><link>https://amfelt.com/blog/why-we-are-building-amfelt/</link><pubDate>Sun, 01 Nov 2026 00:11:00 +0000</pubDate><guid>https://amfelt.com/blog/why-we-are-building-amfelt/</guid><description>A CFO asks the whole company to build with AI, while a modest pilot to collect raw vessel data sits blocked. Why we're building Amfelt for the gap in between.</description></item><item><title>How we keep parallel agents working toward the same plan</title><link>https://amfelt.com/blog/coordinating-parallel-agents/</link><pubDate>Sun, 01 Nov 2026 00:10:00 +0000</pubDate><guid>https://amfelt.com/blog/coordinating-parallel-agents/</guid><description>A commit in August staged one agent session's uncommitted work into another session's change. How we responded with Git worktrees, a guard hook, shared planning documents, a GitHub issue tree, and handovers between sessions.</description></item><item><title>Type casting belongs between the L and the T</title><link>https://amfelt.com/blog/type-casting-is-a-pipeline-step/</link><pubDate>Sun, 01 Nov 2026 00:09:00 +0000</pubDate><guid>https://amfelt.com/blog/type-casting-is-a-pipeline-step/</guid><description>At a wholesaler moving toward dbt, I found typed source columns arriving as strings in Snowflake. The migration showed how much transformation code depended on those conversions.</description></item><item><title>We chose HTMX over React for our app</title><link>https://amfelt.com/blog/htmx-instead-of-react/</link><pubDate>Sun, 01 Nov 2026 00:08:00 +0000</pubDate><guid>https://amfelt.com/blog/htmx-instead-of-react/</guid><description>Why a team with Go and Hugo experience built a CRUD-shaped control plane from server-rendered HTML fragments instead of React, what that costs us, and the kind of app where we would choose differently.</description></item><item><title>We skipped microservices and built a modular monolith</title><link>https://amfelt.com/blog/why-our-platform-is-a-modular-monolith/</link><pubDate>Sun, 01 Nov 2026 00:07:00 +0000</pubDate><guid>https://amfelt.com/blog/why-our-platform-is-a-modular-monolith/</guid><description>Why a small team with a fixed stack chose one Go binary per app on one Postgres instead of microservices, what that costs, and the triggers that would make us split.</description></item><item><title>Keep the site static, put a Go server in front of it</title><link>https://amfelt.com/blog/gating-a-static-site-with-go/</link><pubDate>Sun, 01 Nov 2026 00:06:00 +0000</pubDate><guid>https://amfelt.com/blog/gating-a-static-site-with-go/</guid><description>How we gave a Hugo site per-user access control without giving up static output: a small Gin server, Hanko for sign-in, and the whole site embedded in the binary.</description></item><item><title>Your docs folder is already the knowledge base</title><link>https://amfelt.com/blog/markdown-knowledge-base-for-ai-development/</link><pubDate>Sun, 01 Nov 2026 00:05:00 +0000</pubDate><guid>https://amfelt.com/blog/markdown-knowledge-base-for-ai-development/</guid><description>AI-assisted development produces documentation faster than anyone can file it. Plain Markdown in the repository, rendered by a static site generator, turns that pile into a knowledge base for both people and agents.</description></item><item><title>Your agent writes perfect SQL. It still gets the answer wrong.</title><link>https://amfelt.com/blog/perfect-sql-wrong-answer/</link><pubDate>Sun, 01 Nov 2026 00:04:00 +0000</pubDate><guid>https://amfelt.com/blog/perfect-sql-wrong-answer/</guid><description>Three teams at a national telco counted connections three different ways. That experience, and MotherDuck's data-preparation benchmark, made me question what an agent needs beyond valid SQL.</description></item><item><title>Stop asking AI to guess. Make your pipelines deterministic.</title><link>https://amfelt.com/blog/make-your-pipelines-deterministic/</link><pubDate>Sun, 01 Nov 2026 00:03:00 +0000</pubDate><guid>https://amfelt.com/blog/make-your-pipelines-deterministic/</guid><description>An LLM changed my build check until a failing Hugo build looked successful. That experience shaped how I think about AI-generated pipeline code and the checks it must pass.</description></item><item><title>Every warehouse needs history. Almost none have it.</title><link>https://amfelt.com/blog/every-warehouse-needs-history/</link><pubDate>Sun, 01 Nov 2026 00:02:00 +0000</pubDate><guid>https://amfelt.com/blog/every-warehouse-needs-history/</guid><description>My first engagement as an independent consultant started with a week of manually recovering failed warehouse loads, and a CFO whose numbers moved between Monday and Thursday. The root cause was a warehouse with no history, which is far more common than it should be.</description></item><item><title>I had to read the code to understand the data platform</title><link>https://amfelt.com/blog/understanding-your-data-platform/</link><pubDate>Sun, 01 Nov 2026 00:01:00 +0000</pubDate><guid>https://amfelt.com/blog/understanding-your-data-platform/</guid><description>An incomplete source inventory sent me into years of integration code before I could plan a global warehouse. What that work taught me about documenting a data platform.</description></item><item><title>Data tools are getting better. Data teams are still stuck.</title><link>https://amfelt.com/blog/data-tools-solving-wrong-problems/</link><pubDate>Sun, 01 Nov 2026 00:00:00 +0000</pubDate><guid>https://amfelt.com/blog/data-tools-solving-wrong-problems/</guid><description>Four colleagues praised our demo. The fifth reviewer, a system integrator we barely knew, questioned the complexity and the problems we had chosen to solve. We had to reconsider what would make the platform useful to a data team.</description></item></channel></rss>