If you’re doing breakeven math on subscriptions, consider that your own rig can run 24/7 whereas you will get a fraction of that with sub rate limits. Even if you factor in PG&E residential rates, the breakeven is a lot closer to months for overnight long-running agentic coding a couple times a week.
And in terms of interesting use cases: recently pointed an agent at Blender and gave it vision. That setup can essentially iterate on a scene forever.
I assume this is a strictly B2B proposition in spite of "built for every side of the GPU market" -- meaning I'm probably not going to be able to buy/bid on small (e.g. 2-8) lots of used, matched RTX 3090s for example?
For now, we are focusing on institutions, since most of this hardware has strict export controls. We deal with all sizes though, we can do a single card to 1024 servers. Can always reach out to us to see if a specific request would fit in the marketplace or not
Oddly enough, I did the exact same thing this past weekend, for a couple hundred usd in Fable overage.
It was truly remarkably easy to build and package it exactly to my whims, in my case as a single docker container with a process reaper that runs llama, my Go code, tts, chat harness, browser in xvfb, and even a mailer daemon. With Gemma, it even runs on a RPi 5.
What’s absolutely wild to me is that over a couple hours, I could probably have it import parts of Home Assistant for my devices directly, and do other wacky stuff in what is essentially software for one.
Spent nights over the last six months combining AI coding agents with containers, spec-driven development, and formal verification, so that I (eventually) don't have to manually review 10k+ lines of AI code a day, or worry about the latest model wiping out my home directory:
TL;DR: Overplane is an AI build system that pairs popular AI coding agents with containers, spec driven development, and automatic lightweight formal verification for a safer, replayable AI codegen workflow.
Like many others, I’ve watched AI code pile up at work, and get increasingly lighter reviews due to its size and maddening uniformity. On my side projects at night, I see the full speed and fragility of vibe coding. Neither practice seems sustainable, so I’ve been playing around for the last few months looking for a balanced, middle ground.
The key idea here is similar to chip design: if AI written software functionally does what you intended it to (by virtue of increasingly sophisticated testing against a codification of your intent), then it’s perhaps less important to manually review every line of generated code.
Overplane is a labor of love that brings that idea to life as an open-source experiment, combining some old personal loves: containerization, content addressed build systems, and formal verification.
The example I’d start with is rustdis, a partial, wire-compatible Redis clone in Rust with an empty [dependencies] section. Created with seven short specs and about $45 of Claude Opus, in about three hours: https://www.overplane.dev/examples
What sold me on the approach is that the one IR generated by Overplane from the specs paid for itself downstream: Z3 checks at build time, 48 generated proptest properties, and Kani proofs on the parser and arithmetic core.
It also holds up against redis-benchmark better than I expected, within about 90% of real Redis on my box unpipelined and a bit ahead on some pipelined workloads, which I mostly attribute to rustdis doing less than Redis does.
It’s v0.0.8 and rough. If you’ve tried spec-first workflows or lightweight formal methods in anger, I’d love to hear where they broke down for you.
Such a great comment, and I agree with all of them.
For me in a similar vein:
- mar ‘24: thinking about how to survey the field and implement a hard research task in Natural Language Processing, and then just approximating it well enough with a prompt and a completions api
- mid ‘25: Llama 3 being able to analyze a good sized codebase I was onboarding onto, and synthesize it into diagrams that matched the quality of ones I’d generated by hand with deterministic tools.
- dec ‘25: opus 4.5 basically generating multi-class modules and tests perfectly (syntactically). Finding that errors were my own under-specification of the prompt. Stopped writing code by hand, mainly because it was good enough and came with tests, docs, build scripts, and other goodies for free.
Seems like a similar thought to "Programming as Theory Building"[1].
The paper is well worth reading, but the TL;DR if you don't wanna: the "theory" of a program (how does it fit together? What extensions/modifications would work well, how should it evolve to handle new cases, etc.) is contained only in the mind of the coder; documentation, spoken explanations, etc. are attempts at describing that theory, but they are necessarily a lossy encoding of it. Certainly attrishing everyone who knows something about a codebase would be extremely foolish under Naur's argument.
Debt, I guess to me, is code deviating from a known better theory for it to follow.
The "attrishing" in your comment made me go looking, and Wiktionary [0] says the verb form "attrit(e)" exists and is likely a quite old backformation that has fallen out of common use.
I wouldn't quite say attrite has fallen out of common use, though it is by its nature rarely encountered. I certainly use it, and it's fairly common in certain fields (e.g. discussions of language attrition, ironically).
And in terms of interesting use cases: recently pointed an agent at Blender and gave it vision. That setup can essentially iterate on a scene forever.