The loop in that article is deliberate. ATS exists to externalize the cost of screening onto applicants while giving employers plausible deniability about who gets filtered. The “objective” framing is the fraud. I built a tool for this (CircuitForge-Peregrine, open-core, local-first). Not because I think the system should exist, but because I kept watching people get filtered before any human ever saw their application. A lot of neurodivergent folks especially, people who interview brilliantly but get wrecked by the performance art of keyword optimization. Resume coaches and LinkedIn premium exist for people who can afford to play the game. Peregrine is for everyone else.

Using AI to survive AI screening is genuinely stupid and I don’t love it. But I’d rather the tool be free and local than have the only options be “pay someone” or “lose.”

Working on it! I should have an update later today

Working on it! Will have an update for you later today

Very cool! I’m actually interested in helping with testing and porting to other architecture. Made a comment on the open issue for ARM support, happy to open a PR if you’re interested

Yep, can’t do much about that other than vote and write my useless representatives. Taking back what they stole and putting it to work for the masses, to do the grind, that’s something I can do now.

Peregrine, Snipe, and Kiwi are all available for free on the managed cloud accounts and I’m handing out beta keys for free for a while so please do try them out if you find one that’s useful!

So I am working on the mundane task part…

https://circuitforge.tech/mission

https://circuitforge.tech/roadmap

Basically for fun yes, gamifying development. Anyhoo yeah the idea is you would encounter bugs when discovering bugs or working on bugfix branches, and other types for a few other circumstances.

At the moment the only real rewards are having specific buddies you can assign to specific sessions or threads and they’ll chime in like the OG buddy system, level up based on goals accomplished, no real effect on the code itself, just emergent complexity from whatever the user is up to.

Just FYI, I’m not using LLMs for art apart from placeholders till I can hire someone or find and dust off my old copy of CS6 to start making my own assets. (Inkscape drives me nuts and nobody made a PhotoGIMP-like option for illustrator yet)

All the software I’m designing is deterministic and local first, LLMs and cloud dependencies as fallbacks or gap fillers. And nothing critical without a human to review, etc.