Hacker News: didgeoridoo

New comment by didgeoridoo in "Show HN: Lathe – Use LLMs to learn a new domain, not skip past it"

didgeoridoo — Sun, 07 Jun 2026 18:36:39 +0000

I’m building this! It was originally designed for human accessibility for interactive CLIs, but it turned out to be really useful for giving agents the ability to follow structured workflows.

It runs as a background terminal that the agent can observe, and then exposes all interaction options as structured commands that can be run from the foreground CLI which then update the state of the background terminal via IPC. My hope is to establish a sort of “ARIA for terminals” standard to improve accessibility for both humans and agents. Email in profile, ping me if you’re interested in giving it a spin (just have plugins for Inquirer + Commander right now, hoping to broaden to other frameworks & TUIs soon).

New comment by didgeoridoo in "Ask HN: Best Embedding Models?"

didgeoridoo — Tue, 05 May 2026 10:20:52 +0000

I’m partial to jina.ai — they have open models for code and prose, all easily runnable locally.

New comment by didgeoridoo in "Nuclear receptor 4A1 linked to health effects of coffee: study"

didgeoridoo — Sun, 03 May 2026 16:51:27 +0000

Supercritical CO2 extraction is pretty innocuous. Just buy good decaf from a place that doesn’t bathe their beans in toxic waste.

New comment by didgeoridoo in "Specsmaxxing – On overcoming AI psychosis, and why I write specs in YAML"

didgeoridoo — Sun, 03 May 2026 10:46:32 +0000

I’m building something similar with https://github.com/LabLeaks/special (apologies for the desultory slop-laden README, need to give that a lot more human attention) but I’ve gone in a slightly different direction: a “spec” is a product contract claim supported by attached tests that verify it. It’s a little Cucumber-y, if anyone remembers that, but a lot more flexible — you just write stuff like

  @spec LINT_COMMAND.ORPHAN_VERIFIES

  linter reports blocks that do not attach to a supported owned item.

Then

  #[test]
  // @verifies SPECIAL.LINT_COMMAND.ORPHAN_VERIFIES

  fn rejects_orphan_verifies_blocks() {
    let block = block_with_path("src/example.rs", &["@verifies EXPORT.ORPHAN"]);

    let parsed = parse_current(&block);

    assert!(parsed.verifies.is_empty());
    assert_eq!(parsed.diagnostics.len(), 1);
    assert!(
        parsed.diagnostics[0]
            .message
            .contains("@verifies must attach to the next supported item")
    );

}

And then the CLI command “special specs” pulls your specs and all attached verification + test code so you (or your LLM) to analyze whether the (hopefully passing!) test actually supports the product claim.

There’s also a bunch of other code quality commands and source annotations in there for architectural design & analysis, fuzzy-checking for DRY opportunities, and general codebase health. But on the overall principle, this article is dead-on: when developing with LLMs, your source of truth should be in your code, or at least co-located with it.

New comment by didgeoridoo in "Ramp's Sheets AI Exfiltrates Financials"

didgeoridoo — Thu, 30 Apr 2026 14:55:15 +0000

Amazingly, there is already a recognized verb tense for this: https://en.wikipedia.org/wiki/Prophetic_perfect_tense

New comment by didgeoridoo in "My AI-Assisted Workflow"

didgeoridoo — Wed, 15 Apr 2026 11:34:28 +0000

There is no evidence of this. Evals are quite different from "self-evals". The only robust way of determining if LLM instructions are "good" is to run them through the intended model lots of times and see if you consistently get the result you want. Asking the model if the instructions are good shows a very deep misunderstanding of how LLMs work.

New comment by didgeoridoo in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

didgeoridoo — Mon, 06 Apr 2026 19:09:40 +0000

Haha no that’s change - 4.4x MORE expletives per word in the last week.

New comment by didgeoridoo in "Claude Code is unusable for complex engineering tasks with the Feb updates"

didgeoridoo — Mon, 06 Apr 2026 17:48:46 +0000

There are indeed non-expletive words that can contribute to the denominator, though I use them less and less these days.

New comment by didgeoridoo in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

didgeoridoo — Mon, 06 Apr 2026 17:06:59 +0000

Running some quick analysis against my .claude jsonl files, comparing the last 7 days against the prior 21:

- expletives per message: 2.1x

- messages with expletives: 2.2x

- expletives per word: 4.4x(!)

- messages >50% ALL CAPS: 2.5x

Either the model has degraded, or my patience has.

Show HN: Claudebar, the missing interactive menu bar for Claude Code

didgeoridoo — Fri, 03 Apr 2026 01:11:37 +0000

claudebar wraps Claude Code in a detachable, resumable tmux session with a nice status bar (showing things like session usage % and the new Peak vs. Off-Peak status), interactive side panels to manage Tasks and Agent Teams, and a clickable menu for when you're feeling mousey. I've found it to be a useful little enhancement to my everyday work. Hopefully you do too!

Comments URL: https://news.ycombinator.com/item?id=47622239

Points: 3

# Comments: 0

New comment by didgeoridoo in "The widely reported "hole in the Universe" is a lie"

didgeoridoo — Thu, 26 Mar 2026 12:48:10 +0000

Thousands of words to say:

- cosmic voids are real but not actually empty, just lower density than average

- pop science articles sometimes mistakenly use pictures of Bok globules to represent voids

This is probably one of the lowest signal-to-noise ratios I have ever seen.

New comment by didgeoridoo in "Show HN: Agent Kernel – Three Markdown files that make any AI agent stateful"

didgeoridoo — Mon, 23 Mar 2026 22:18:39 +0000

Along these lines, I’m working on a tool called Spotless[0] that takes a more HTTP proxy-based approach to make statefulness something the agent doesn’t have to worry about. It directly reads & writes to the messages array going to and from Anthropic, so you don’t have to rely on the agent calling an MCP or using a skill. Still buggy and early, but it’s definitely a very interesting way of working with agents.

https://github.com/LabLeaks/spotless

New comment by didgeoridoo in "Metacompiler – A Novel by Michael Barr"

didgeoridoo — Wed, 18 Mar 2026 01:10:55 +0000

There’s also the excellent Coding Machines (2009): https://www.teamten.com/lawrence/writings/coding-machines/

New comment by didgeoridoo in "Throwing away 18 months of code and starting over"

didgeoridoo — Tue, 10 Mar 2026 19:48:30 +0000

That and Brooks’ underrated “The Design of Design” are notable for having an almost impossible density of quotable aphorisms on every page. They’re all so relevant today that it’s hard to believe that he’s talking about problems he faced half a century ago.

New comment by didgeoridoo in "You Hired the AI to Write the Tests. Of Course They Pass"

didgeoridoo — Tue, 10 Mar 2026 19:45:49 +0000

Isn’t this just an API sandbox? Many services have a test/sandbox mode. I do wish they were more common outside of fintech.

New comment by didgeoridoo in "Google PM open-sources Always On Memory Agent, ditching vector databases"

didgeoridoo — Sat, 07 Mar 2026 21:06:20 +0000

I just built something similar, specific to Claude Code. It runs as a transparent HTTP proxy that reads & rewrites the entire messages array that CC sends to its API. Same “dreaming” consolidation approach (using Haiku and another instance of CC itself, so it uses your subscription). Check it out!

https://github.com/LabLeaks/spotless

New comment by didgeoridoo in "Bourdieu's theory of taste: a grumbling abrégé (2023)"

didgeoridoo — Sat, 07 Mar 2026 18:12:32 +0000

> lower-class people are in a sort of local maxima

If the writer knew that the correct term is “maximum” (singular) and misused the Latin on purpose, this is brilliant. Failing that, it’s still a wonderful inadvertent enactment of the thesis. Well done either way.

Dereks at Work: what would it mean for an AI agent to be "accountable"?

didgeoridoo — Thu, 05 Mar 2026 20:39:47 +0000

Article URL: https://www.lableaks.dev/p/dereks-at-work

Comments URL: https://news.ycombinator.com/item?id=47266984

Points: 3

# Comments: 1

New comment by didgeoridoo in "Agentic Engineering Patterns"

didgeoridoo — Wed, 04 Mar 2026 19:06:37 +0000

I don’t know, Simon has had a pretty sane and level head on his shoulders on this stuff. To my mind he’s earned the right to be taken seriously when talking about approaches he has found helpful.

New comment by didgeoridoo in "Saturday Night Live mocking people with disabilities"

didgeoridoo — Mon, 02 Mar 2026 11:32:41 +0000

If someone with coprolalia is involuntarily induced to say the most awful, inappropriate thing possible in a given situation, doesn’t shouting slurs show that they aren’t racist?

Someone who is deeply racist would presumably consider racial slurs to be neutral statements, and not actually care about who they are offending. I wonder if that would actually steer the coprolalia away from racial slurs and toward something else that one has internalized as truly “offensive”.