to some degree, but this is also kind of like saying most restaurants are just interfaces over calories
sometimes everything around the product is the actual product
most sandbox products are just interfaces over microVMs, almost like SSH or SMTP
everyone is treating the interface as the product, while the interface is more like an open protocol and the product is the runtime underneath
in 2016, the sandbox interface would have been a CNCF project - a standard interface for `vm new` and `vm stop` with swappable layers for Cloud Hypervisor, Firecracker, and QEMU
but because it's 2026 and everyone can ship code effortlessly, we get 100 companies trying to commercialize the same interface instead of having one open standard
it’s a challenge to know what’s worth addressing and what’s just unnecessary complexity…Claude will happily find more and more issues if you keep asking it to
it feels like the job of the human is to defend the codebase from slop for as long as possible, but sometimes it’s hard to decide what is and isn’t slop!
sandbox cpu-hours just aren't fungible like this afaik
shared vs dedicated, different cpu generations
it's more helpful to have a task-based benchmark that compares time to completion + total cost. depending on the task, that may also include network, memory, disk, etc
Every document parser is #1 on its own benchmark.
So forget static benchmarks. Today, we're releasing ParseJudge, an open-source app that judges parsers on your own documents, free for up to 10,000 pages a day.
https://t.co/sDYDNsmgDW
@mjackson I think this has been the big transition of 2026 - people kinda just stopped mentioning “vibe coding” cause that’s just how so much gets built now 😳
I think using embeddings with “traditional ML” is really cool and maybe under explored (or at least under hyped) for unlocking a different and cheaper kind of intelligence
EmbeddingGemma 2 maps text, images, video, audio, and code, running 100% locally on your device.
At just 740M parameters, you can build powerful multimodal apps running entirely in your browser via WebGPU & Transformers.js. From multimodal search to cross-modal classification and clustering, local embeddings unlock endless possibilities.
It's been a while since I showed any Jev stuff but I've been cooking on a way to use @GoogleDeepMind's EmbeddingGemma2 as an insanely fast real-time /v1/systemone endpoint (20ms on text, 100ms on images).
It's trainable in real-time, can distill an upstream teacher model, and is multi-modal by default. Hard to believe a @googlegemma model of this size can do so well with the right harness!
@Neriousy@opencode very cool. I'm trying some long-running @opencode sessions in goal mode to see what it can do. made a CLI to run them in @sailresearch sandboxes
https://t.co/ROLnoGzvWn
on @PowersetRes we run an LLM embedding pipeline to categorize repos based on their READMEs and website content.
we use embedding similarity to a category centroid to determine membership- it works, but seems like a place Jev could really shine
could simplify a whole lot and probably be faster/cheaper