Authentication Stuff:
hnchat.com:0m9YzMwf4kdLeEnrHt2n
547e7b5026bc47ac8302bf3d213909be
[ my public key: https://keybase.io/supermdguy; my proof: https://keybase.io/supermdguy/sigs/fHS654nzPCSpfpoOPNrHUJrgc2jOy21zNsnFwyeLDfs ]
> We're also launching GPT‑5.6 Sol on Cerebras at up to 750 tokens per second in July, bringing frontier intelligence to customers at unprecedented speed.
This is really exciting. I work on voice AI, and we're still using 4.1/4.1 mini since none of the frontier models come close on latency. I'm excited to be able to have more interactive experiences, I think it'll unlock new ways of working with these models.
One trick I’ve used is creating a folder and then adding a .gitignore inside it with *. Then nothing in that folder gets tracked, without needing to add anything to the public gitignore. Didn’t know about .git/config though!
I agree that current memory systems are pretty bad, and I think that’s because memory is a prompted behavior instead of a learned one. In theory, if memory was an emergent behavior instead of a prompted one, it would be a lot better.
I think you’re right that changing its own harness would be bad and skew towards prompt engineering instead of learned behavior. So maybe instead it could start with a harness with memory CRUD tools and then learn how to most effectively use them.
There was a show HN for something similar a couple months ago[0]. Looks like they shut it down. Probably too difficult/low-margin to run as a business, but I think the co-op model you mentioned has potential.
It's interesting because their last model series (Phi) was based around the thesis that high-quality synthetic data is better than a large pre-training corpus.
I think the conduit exception still applies for analog faxes. Which makes no sense, since tapping a fax line is probably way easier than compromising a data center.
Apparently from a third party seller in New York....and tastes really bad. I was surprised steak could be safely mailed in such normal looking packaging!
Interesting to see this after the recent post about Chrome’s on-device model using up 4gb of storage, which frustrated a lot of people [1].
I agree local models are great, and it’s cool that Apple has models built in now. But I feel like it basically has to be an OS level feature or users are going to get upset. I’d certainly rather have a small utility call out to OpenAI than download its own model.
Better auth is great! I love how it's way more hackable than using a something like Clerk. We were able to add a plugin to allow auth via iframe postMessage (embedded in a CRM) and everything worked seamlessly.
I've used it off and on over the last month or so. For more complicated tasks (30+ minutes) it works well, and seems to replace a lot of prompting that I'd normally need to do (e.g. asking questions about requirements, creating specs and implementation plans, staying on task). For simple tasks, it tries to do too much and gets in the way.
> The Yes people are betting that, later this year, their counterparties (the No betters) will want cash (to bet on other markets), and so will sell out of their No positions at a higher price.
Overall, I'm really impressed by what you accomplished! I'm not a researcher, so not sure if this is that helpful, but here are some thoughts:
- I wonder if the "move" action is difficult for the model to learn to use well. The model sees token location as positional encodings in the embedding, not sparse character offsets. Would be interesting to see something more like "jump to next/previous [token or set of tokens]". Or maybe a find/replace like most coding harness edit tools use?
- I'd move the exact training data generation details to an appendix. Could be summarized to improve the flow of the paper.
Cool concept! I think the hardest part will be getting people in the target audience to use it. A lot of indie hackers make software for other indie hackers, but that isn't true of most other verticals. And honestly building software for indie hackers feels like a losing battle. Any ideas of how to incentivize none-builders to rank projects?
From the most recent comment, looks like this is a bug, triggered by the system inadvertently activating an internal release tool [0]. Still a pretty wild bug, but not as dramatic as the title suggests. Which is kind of unfortunate honestly, the chaos of every gas town instance automatically contributing to itself would be beautiful to see.
Authentication Stuff: hnchat.com:0m9YzMwf4kdLeEnrHt2n 547e7b5026bc47ac8302bf3d213909be [ my public key: https://keybase.io/supermdguy; my proof: https://keybase.io/supermdguy/sigs/fHS654nzPCSpfpoOPNrHUJrgc2jOy21zNsnFwyeLDfs ]