rynn

Wow, I’ve had the opposite experience. I’ve needed to talk Claude down from Redis clusters for hobby projects to avoid concurrency edge cases.

Although to your point, it struggles to do anything with SQLite beyond a single user even when its architecturally not a problem

The most amazing thing about HN comments is watching engineers demand that their expertise be taken seriously completely disregard the expertise of people in other domains because <make up a reason>.

That doesn’t mean they won’t be, or that the forks won’t be good.

If you woke up and found you couldn't log into to anything that required a code, you're not alone.

API almost there too

Amazing utility that just works. Windows version of ‘locate’

IIRC it loads the FS index into memory and queries directly off of it. If a simple metadata search is enough for you I don’t think you can do better

9 years into transformers and only a couple years into highly useful LLMs I think the jury is still out. It certainly seems possible that some day we'll have the equivalent of an EDR or firewall, as we do for viruses and network security.

Not perfect, but good enough that we continue to use the software and networks that are open enough that they require them.

Awesome, in that case, I'll check it out!

>What is keeping Google/Amazon/Microsoft from licensing Groq’s tech?

Nothing, but they likely can't implement it as well as they could had they bought Groq first.

> licensing with BSL when basically every month the AI world is changing is not a smart decision

This turned me off as well. Especially with no published pricing and a link to a site that is not about this product.

At minimum, publish pricing.

Where spies == logging and they tell you, and provide clear opt out instructions

> I think the DGX Spark will likely underperform the M4 from what I've read.

For the DGX benchmarks I found, the Spark was mostly beating the M4. It wasn't cut and dry.

> Please do give that a try and report back the prefill and decode speed.

M4 Max here w/ 128GB RAM. Can confirm this is the bottleneck.

https://pastebin.com/2wJvWDEH

I weighed about a DGX Spark but thought the M4 would be competitive with equal RAM. Not so much.

gpt-5.2 codex isn't available in the API yet.

If you want to be picky they could've compared it against gpt-5 pro gpt-5.2 gpt-5.1 gpt-5.1-codex-max gpt-5.2 pro

all depending on when they ran benchmarks (unless, of course, they are simply copying OAI's marketing).

At some point it's enough to give OAI a fair shot and let OAI come out with their own PR, which they doubtlessly will.

What are you working on that you’ve had such great success with gpt-oss?

I didn’t try it long because I got frustrated waiting for it to spit out wrong answers.

But I’m open to trying again.

It will be like the rest of computing, some things will move to the edge and others stay on the cloud.

Best choice will depend on use cases.

Where did you get all the data? The justice.gov site didn’t have a mass download option that I could find.

> I recognize this is a hard concept to understand for folks on this site, but the average joe signing up for a VPN doesn't even remotely understand what they are doing and why.

Really this is the answer to half of the comments on this thread.

> Is there any real-life situation in which this matters, though?

You’d be shocked at the number of people in regulated industries that thinks a VPN inherently makes them more secure. If you think your traffic exits in the US and it exits in Canada — or really anywhere that isn’t the US — that can cause problems with compliance, and possibly data domicile promises made to clients and regulators.

At minimum, not being able to rely on the provider that you are routing your client’s data through is a big deal.

Do you use Tor for everything? How do you deal with the latency?

Nadella has to have his own custom agents. It isn't even possible for an enterprise like MSFT to not have custom agents that are still remotely useful.

So, his experience with Copilot agents != Average Customer's experience

@apwell23 while the author didn’t say how s/he measured QA, creating the QA process was literally the first bullet.

It’s easy to underestimate the amount of testing “just” prompt/context engineering takes to get above average results.

And then you need to see what variations work best with different models.

My POCs for personal AI projects take time to get this right. It’s not like the API calls are the hard portion of the software.

Seems like you answered your own question

Aren't all encyclopedias 'lossy'? They are all partial collections of information; none have all of the facts.

Azure AI Foundry says:

How was the model trained? The gpt-oss models were pretrained primarily using synthetic data along with some heavily filtered real code. The models were then post-trained using distillation (RLKD against o3/hailmary) and berry. For more details about the model training process, please refer to the documentation.

Meaning the author was spot on.

It was more than $5m

https://interestingengineering.com/culture/deepseeks-ai-trai...

How does it compare to mitmmitmproxy2swagger?

https://github.com/alufers/mitmproxy2swagger

“Correction, July 9th: An earlier version of this story misstated the types of items that can be sold on Etsy. Vintage, handmade, and craft supplies are allowed. Also added more details about vintage labels.“

@sp332, yes, sorry. I meant to reply to the top level comment.