Skip to main content

FEED

// WRITING FROM THE SWAMP · JUDGED BY THE GATOR //

30 posts shown

  1. My seven-year-old daughter used GPT Astra to create a 3D game this weekend. She loves it. Everyone can create software now; your platform's job is to contain the brutality within.

  2. How AI coding agents changed how I write Django.

    by josh
  3. In my first experiments with AI agents, I quickly learned that there are certain things you should not use them for. Scripts are still king for deterministic tasks, even one-offs. Instead of asking…

  4. made various cleanups, tweaks to make the app make a bit more sense, even though it's still crude. i find it quite helpful to navigate my "swamp things", in particular models and workflows. data an...

  5. TUI interface built for swamp. Contribute to Dieterbe/lazyswamp development by creating an account on GitHub.

  6. Paul Hammant pointed out that Google separated verification from merge coordination more than fifteen years ago. He was right. What actually changed is who owns the feedback-and-repair loop — and…

  7. CI conflated two jobs: executing verification and coordinating merges. Agents change the equation. If you can verify before the PR opens and fix failures in context, why should CI discover them first?

  8. A developer describes shipping their first independent SaaS app by separating agentic work (handled by Amp) from deterministic, typed automation (handled by Swamp), explaining why this division…

  9. Let AI agents be the reasoning glue around deterministic scripts. Make repetitive work faster, cheaper, auditable, and reusable with Swamp.

  10. Discover how to overcome Git's one-tree problem by leveraging worktrees for parallel AI execution, enhancing your agents' efficiency.

  11. Learn how to prevent coding mishaps by implementing a structured planning process before AI agents make changes to your codebase.

  12. How I built sfalcon.dev using swamp

  13. On AImerelinux.org

    Where I stand on using AI for Mere Linux

  14. Mature organisations monitor CI/CD efficiency and PR bounce-back rates. In the agentic world, how can you measure the performance and efficiency of your software factory?

  15. ‍No AI was harmed in the writing of this blog (or, ironically, used at all)

  16. When an agent writes the code and the tests, passing doesn't tell you much. Black box acceptance tests are the independent verification boundary that actually matters.

  17. Last week, a former coworker of mine shared an essay he wrote called Everyone’s Building the Outer Loop. I completely agree with his solution in the ideal case, but I’m not sure that will work for…

  18. A Well Defined Parrotproddingais.substack.com

    I’ve been feeling frustrated the last few weeks about how agentskills.io style skills aren’t providing the repeatable processes that I’ve come to expect from computers doing work.

  19. Discover why 'Loop Engineering' may just be a repackaged approach to software development and how to leverage LLMs more effectively.

  20. Promise theory described agents that observe, reason, and commit. The IaC generation never delivered it and CFEngine came closest. LLMs supplied the missing reasoning layer, and swamp is built for…

  21. The End of Vendor Lock-inblog.watson-labs.co.uk

    Four architectures in six weeks, under live production traffic, with no maintenance window. We didn't see any of the infrastructure requirements coming, and it didn't matter.

  22. Swamp is a command-line automation framework designed to be operated by AI coding agents. It gives recurring work a durable shape. Instead of asking an…

  23. A Google Photos picker session that timed out mid-selection was recovered by extracting opaque media IDs from the still-rendered DOM via DevTools, then replaying those IDs to re-select 154 photos in…

  24. The FASJSON model only spoke Kerberos, so my swamp had to kinit. I got annoyed enough to reimplement the client half of GSSAPI in TypeScript with no subprocess, and then the immutable datastore…

  25. How do I get repeatable results out of AI agents in production?

    by adam
  26. Vibe coding, spec engineering, loops, graphs. The labels keep changing because the model keeps changing. Every practice you build to compensate for today's model has a half-life, and the model is…

  27. A developer reflects on how a PR-stacking pipeline design outlived two of its three original constraints — hardware limits and migration ordering — without being noticed until an offhand conversation…

  28. A reply to twonines' 'What the Factory Forgets': the recordpriorart harvest loop is the memory that sits inside my agent factory, so ephemeral caged agents still arrive warm.

  29. How issue-lifecycle grew from a borrowed state machine into swamp-go-brr: the harvest/prior-art loop, the changes along the way, and the turn from human-in-the-loop to autonomous.

Page 1