rybosome

12 ms·

121.
by rybosome 2y ago
This doesn’t make LLMs worthless, you just need to structure your processes around fallibility. Much like a well designed release pipeline is built with the expectation that devs will write bugs that shouldn’t ship.
122.
by rybosome 2y ago
Fascinating. R1 really punches above its weight with respect to cost-per-token. As the article alluded to at the end, my thoughts immediately go to using R1 as a data generator for complex problems, since we have many examples of successful
123.
by rybosome 2y ago
Very much agreed. I spent a decade in FAANG, then burned out and left. I’m at an early-stage startup now and couldn’t be happier. The pressures are very different. My most stressful episodes at this startup have been related to technical pr
124.
by rybosome 2y ago
Thanks for sharing this. I find the hostility in the comments bizarre. Do people really think that AI consciousness is impossible? Or, that if it is possible, it will be obvious when it emerges? To be clear I still doubt we are there yet, b
125.
by rybosome 2y ago
Yes, our previous lead dev built a lot of our infra using LangGraph. I've been slowly ripping it out since assuming ownership of this part of the codebase. I've been replacing LangGraph with simple primitives, relying on native Py
126.
by rybosome 2y ago
I felt deeply vindicated by their assessment of these frameworks, in particular LangChain. I've built and/or worked on a few different LLM-based workflows, and LangChain definitely makes things worse in my opinion. What it boils d
127.
by rybosome 2y ago
Yeah, you’re right that it’s not a strong sell. It’s not clear to me why I’d want to use this over any other more mature framework.
128.
by rybosome 2y ago
FastAPI also supports dependency injection.
129.
by rybosome 2y ago
People have tried that. Supposedly, their drones are disabled when they get close.
130.
by rybosome 2y ago
I can't comment on why the model gave you that code, but I can tell you why it was not correct. `queue.remove(0)` gives you an `int[]`, which is also what you were assigning to `current`. So logically it's a single element, not an
131.
by rybosome 2y ago
I’m not going to shame you for your parental experience, but it does not mirror mine as a father. When my daughter was born I was crying with joy. And while her infancy was enormously challenging, especially as she was born right when COVID
132.
by rybosome 2y ago
I once read an article that speculated about this. The conclusion was that perhaps there could be a ring around the planet, at the border of the side facing the sun from the side facing away from the sun. That ring could be a habitable zone
133.
by rybosome 2y ago
I suppose I have perfect pitch, in that I can more often than not: * aurally identify key signatures * aurally identify chords * sing a given note on command It wasn't until I was into my early twenties that I could do this.
134.
by rybosome 2y ago
Perhaps debris from an asteroid impact.
135.
by rybosome 2y ago
That’s funny, but open weight, local models are pretty usable depending on the task.
136.
by rybosome 2y ago
> Hot loading is clearly better than anything else when you've got a million clients connected and you want to make a code change. In the contexts in which I’ve worked, this was solved by issuing a command to the server to enter a l
137.
by rybosome 2y ago
I’d love to see what performance benchmarks look like. The old ones were highly optimized, but perhaps for different challenges than today’s architectures present.
138.
by rybosome 2y ago
I was able to determine what this is, and I’m not a specialized scientist. I merely exercised the discipline to read the post before reacting and commenting.
139.
by rybosome 2y ago
The project this “cloned” is just a data visualization tool. You can plug any data into it - good data, bad data. They are launching an AI model which they claim produces higher quality weather data than traditional models relying on physic
140.
by rybosome 2y ago
I suggest you read the post. Reading is typically how information is transmitted. EDIT: the post I am responding to was altered to sound much less confrontational. It was originally: > So what exactly are you “launching” and why does it
141.
by rybosome 2y ago
You went to the effort of posting 2 HN comments, tweeting, and taking a screenshot because you are annoyed that this project used an open-source library and were transparent about doing so.
142.
by rybosome 2y ago
> We visualize these using [cambecc/earth] ( https://github.com/cambecc/earth ), one of our favorite open source weather visualization tools From the post.
143.
by rybosome 2y ago
I take it to mean less pleasure, which I suppose is a form of suffering. I don’t follow the logic of the comment you’re replying to that degrowth would be a hard sell to a democracy, but reducing comfort and pleasure wouldn’t.
144.
by rybosome 2y ago
The poster answered your question. The game is unplayable without an internet connection - it refuses to start. In spite of the fact that the internet connection is only used to provide optional content, without an authenticating server thi
145.
by rybosome 2y ago
Unsurprising outcome. I got a chance to try one at a former coworker’s house. It’s an impressive piece of tech; the immersive video was amazing, felt orders of magnitude more impactful than the SD->HD leap. I also found the eye tracking
146.
by rybosome 3y ago
Generally I agree with your point, but there is some value in a parser that doesn’t have to be updated when the underlying HTML changes. Whether or not this benefit outweighs the significant problems (cost, speed, accuracy and determinism)
147.
by rybosome 3y ago
Thanks for the insights. Any particular resources you’d recommend to learn more about this? So if I’m understanding it correctly, there’s one consistent way that Ollama will vectorize a set of text. Perhaps there are various ways one could,
148.
by rybosome 3y ago
> Your python script is ephemeral and run on demand. You don't want to load the model each execution because it's slow, so you need a daemon. Ollama can serve that role without you having to write anything other than the script
149.
by rybosome 3y ago
Thanks. I wish it was better documented, have a ton of questions about how it functions in practice that I’ll end up trying to reverse engineer. Like, what’s the lifetime of this context? If I load a new model into memory then reload the or
150.
by rybosome 3y ago
Since you brought up that context, do you happen to know how it works? I tested it, and it definitely works like a history - eg you can feed it to a model and ask it a question about what you just talked about, and it works. Does Ollama ess

More ›