fhub

Sophisticated? Perhaps. But could also be run of the mill honey trap/romantic approach. Easier to slip by 2fas through a front door.

18 year old endorsements are cute touch.

Human override could have 2 keys that get turned concurrently at two locations by people who never meet each other, never talk, each has different job incentives.

The glass half full take is that we are very lucky that there is some much literature that pre-dates LLMs. Our brains are tuned to enjoy that more. People who grow up with LLM slop from the start won't necessarily see it the same way.

Casually insinuating serious sexual misconduct about someone without evidence seems pretty far outside the spirit of HN.

Are you insinuating something about David Ellison here?

When I used to ride around SF on an e-bike, I’d deliberately choose undulating routes so I could climb at max assist speed and then fly down the descents above max assist speed.

Compared to who? Every professional developer? Someone learning to code? Fabrice Bellard?

I relate to this. My extravert wife throws the parties. I’m the introvert. I enjoy parties at our place so much more than at other peoples.

My father is 75. He often says with a smile, “Let’s just Google AI it.” The smile is because he thinks he’s cutting edge. He hasn’t been cutting edge for 30 years, but Google has made AI so accessible through his existing search habit.

I imagine their retention data through this uncanny valley transition is very good.

Peak throughput per day is 3k req/m.

I’m in the same boat. Hate typing on a phone. But due to chat interfaces,I’ve learned to mutter discretely into Apple keyboard dictation.

I code and maintain a Rails stack. About 20% of the time an endpoint takes to respond is spent in Ruby. New Relic tells me our Apdex is 99, and it’s very rare for an endpoint to take more than 100ms. Most respond in under 80ms. I could LLM-port it (and probably will one day), but end-user response performance wouldn’t be the motivation.

What does make me consider switching is reading about things like how Shopify’s native apps are moving towards superfast product cores that can be tested incredibly quickly, with the slower UX layer kept separate. I think that model is the future, and in that world Rails starts to look pretty dead.

Was this another case of a pentesting agent breaking out of its sandbox and accessing the open internet?

It would be interesting to hear more about how you’re actually using them. Do you have sophisticated feedback loops around the models so they can verify their work and converge on good solutions? And how do you decide what to give the 5090 vs the Spark vs a frontier model?

Correctness matters much more than speed to me, but if I can get both, that’s obviously very interesting.

How are you deciding which work to send to the 5090 vs a frontier model, or making the two work together nicely?

Correct is much more important than fast for me, but if I could get correct and fast, that would obviously be amazing.

Context

“I’m thinking about banning Claude Code at Shopify until they change their mind and read AGENTS.md and .agents/skills etc.,”

- Tobi Lütke on X.

Cause: Legacy Salesforce login service got into a resource-exhaustion cascade.

Fix: Rolling some unspecified fix they proved in testing out over the fleet seemingly very slowly (After their earlier attempts to roll something out faster failed).

Details at https://status.salesforce.com/incidents/20004433

If you can get your agent of choice to run up the app in an emulator and navigate the whole thing taking screenshots then tasks like edge-to-edge can be pretty hands off.

The way I did it was to firstly get accessibility identifiers on everything. Then have it record everything I do on an emulator to navigate the entire app. Then left it for a while to work out how to actually use the app via the emulator. Once you have this sort of baseline. A lot of changes are really quite pleasant. I often ask for an HTML gallery of screenshots covering a few device sizes as the step before I bother running it on a device myself.

Some reviews are clearly automated, some get a light human review, and some involve a human really looking hard. New apps seem to get a more rigorous human review for their first 2-4 submissions. Or at least humans looking over automated test screenshots etc. One reviewer apparently only looked at screenshots because they missed something that would be blindingly obvious on the prior screen if they did it themselves.

The explosion in LLM app development has clearly created a bottleneck at the human review steps. Not only delays but much more “dumb” rejections from likely over-stressed humans.

FWIW, I find if I’m pushing the disk limit, Spotlight search becomes unusable. So I free up ~100GB (easy to do with iOS/Android dev artifacts, emulators, etc.) and then force a Spotlight re-index from the command line:

sudo mdutil -E /

You can check indexing status with:

mdutil -s /

After that it’s usually fine again until I start pushing the disk limit.

Every sibling comment's issues I've experienced. But a reindex after freeing enough disk space seems to be a semi-stable solution for me. If only I could write a tool to remind me to free disk space before it screws up spotlight. It would be nice if MacOS put the index on its own partition.

Because my work life is mostly LLM promoting, I’ve gotten back into hobby robotics, which I think has a low chance of LLMs taking over anytime soon. Designing and building a robot that can build a wooden hardwood deck using the edge-screw method.

The comments here are taking a very uncharitable definition of “power”. If I take that definition I can see how people are being so negative. But just a slightly more positive definition and this all seems like reasonable dialogue to have with an early stage startup.

“LG TVs process voice data only when the voice button on the remote control is pressed and held, or when a wake word such as 'Hi LG' is recognized after the user has activated the Far-Field voice recognition feature."

The definition of “process” is really important and is missing. LG could process all transcripts on their servers too and stay within a very wonky definition of “process” on the TV that excludes transcription and transmission.

I had one feature in RN and the rest properly native. The day I realized agents were good, I ported that RN to native. Monkey-off-back moment.

I think for OpenAI to win back some hearts and minds here we should have the option to retrospectively turn off "Help improve our AI models". i.e. Any new model trained would exclude all those user's sessions. This could be technically hard but I'm sure an intelligent AI model could work out how to do it :-)

ChatGPT agrees with this too.

https://chatgpt.com/share/6aa31959-b0e8-83ec-bee6-851ed18d45...

She turned herself in (After the police went to her home and she wasn't there)

Package size has very low correlation to launch time. The binary and assets are separate. Additionally some slow app launches are simply waiting for a network request or chain before the view controller renders.

We need to be building silos to save humanity. Maybe 50 of them should do it.

1. Onboarding tells you to

2. Convenience

3. Cheaper than buying an Apple TV et al