noemit

whats wild is there are people who look very fat - who have lots of subcutaneous fat, who don't have visceral fat. The excessive subcutaneous fat can be hard on your joints but doesn't seem to correlate to other health issues.

strangely addictive

They've already done this through the 6m Anthropic API "unlimited plan" grants. Claude-only is getting embedded.

i think its more efficient for models to write queries than read large(ish) documents. at some point you have to have a tradeoff - smaller docs, but the model has to find them (thats a query) or larger docs, and the model has to read them.

My take is AI native apps that have longevity will prioritize token-efficiency - and I believe queries do that.

Does it record the audio as well? Asking for my lawyer.

I assume it was too expensive, because it's really not a bad tool. I used it recently to make my twitter pfp :)

Theres tons of ai apps. They're all general use chatbots or coding agents. Manus, Cursor, ChatGPT. Almost every app that has a robust search uses a reranker llm. AI is everywhere.

As far as totally new products - I built one (Habit.am - wordless journaling for mental health) and new products require new habits, people trying new things, its not that easy to change people's behavior. It would be much easier for me to sell my little app if it was a literal plain old journal.

Even if it runs, this will run slowly, and heat up.

I think local will always have a place, but the infrastructure is going to be used in my humble opinion.

is it just me or could every set of claude skills just be a blog? I don't install skills without reading them first anyway.

I ran into it while building - I should have tested different temps too - I was just trying to get cli style tool calls to be more reliable

I ran tests of 100 attempts with different prompt/scenario combinations. Each "attempt"/theory had 3 different system prompts wordings. Most of the prompts did not mention a colon, but it kept appearing. When I added negative instructions against using a colon, the quality went down (most of the tool calls were malformed, one common issue was markdown ticks in front) It was only when my system prompt acted like colons were normal that I kept getting 100/100 perfect expected tool calls. I ranked my system prompts by which returned the most consistent commands.

I also expected this. Please run some experiments and maybe other models are different

I tried over 20 variations of different system prompts. Once I changed my tool to expect the colon, it also felt like it was running/calling tools faster, but I need to do a larger test to be sure.

Try this prompt "Hi Claude, please make my site accessible. Pay particular attention to contrast ratios for text and the backgrounds they are on.

I dislike that the source of the skill is not linked directly.

Your best paid engineer is the one who reviews what all your token-burners are creating.

I agree. Kimi K2.5 is basically a clone. But also GPT 5.4 has replaced Opus for me. I have unlimited access to all of the proprietary models so I really only make the choice based on convenience.

How are open weight models so much cheaper then?

I must be alone in this but I don't think its heavily subsidized. I see their models as really overpriced. No way they cost that much. Could they really?

Dario has stated multiple times he doesn't believe there is any value in open-weight models. Not surprised. This is not the behavior of an innovative company - it is fear-driven. They are seeing a rapidly shrinking moat.

I canceled my family subscription last month. We have Youtube Premium which I use to play music - Spotify was over 20/month and no longer made sense. We mostly used it to play music on our Google speakers for the kids. It's one of those products that used to make sense, but now just doesn't feel so critical. It's easy to replace and patch where I used it.

He's had tons of "new companies" since Uber.

Progressive Web Apps Exist. You can download any app. And build them too. I build my own apps that send me notifications from my AI Buddy :)

Try saving my side project to your home screen : Habit.am - works really nicely once you're logged in.

is the motto, "Don't be good?"

This is actually sick AF and I found it useful.

agree. I would never do what OP did. But I also won't throw out my smart watch (Context: other people in the thread said they stopped using them because the data was useless)

Funny: Torrenting is (for personal use) legal in Spain, streaming IP (in a local pub) is not.

A counter example:

I've been wearing an Apple Watch for close to 10 years. I've tracked my weight as well along those years but nothing crazy like OP. The Apple watch tracked plenty.

I had some strange symptoms and two doctors insisted I had a weak heart and potential heart failure. This was shocking! Turns out I do have a really "weak" rhythm, but heart failure is when your heart is progressively getting worse in it's pumping. I don't even remember which metric he looked at in my Apple health - but basically my heart has always been this way. A doctor looking at a single data point might think I have abnormally low blood pressure/heart rate, but if I've had this for 10 years with no change, the medical assessment is very different - it means nothing. Sometimes boring data is exactly what you need. For this reason, I will probably always wear an Apple watch (or equivalent) moving forward.

Data can feel useless for 10 years until one day it becomes critical. The benefit is spiky and uneven.