Viewing profile — pedrovhb

Very interesting, thanks for sharing. I'm curious about the focus on the language design and features with regards to agent orchestration vs. being a general purpose/ML architectur…

Nice! Seems very useful if you can drop in and have everything work. Nitpicking a bit: it's not as much _rendering_ markdown as it's _syntax highlighting_ it. Another interesting a…

You can presumably run code that calls `sys.settrace` for that. Which makes it somewhat underwhelming to realize that you pretty much could also do that before, but perhaps conveni…

As a Brazilian - Pix was a pleasant surprise, especially in that for once it feels like we're not lagging behind. It's convenient, free, instant transfers across banks. You can als…

> but most probably was training data to prevent deepseek from completing such prompts, evidenced by the `"finish_reason":"stop"` included in the span attributes As I understand, t…

I imagine it wouldn't necessarily require their opening of remote connections, just a misconfigured reverse proxy.

This seems centralized, though you can self-host it.

> [the extensive anti-reverse engineering measures are] more annoying than any financial app I've had, and I have 5 of them on my phone Ah, this reminds me of the Tuya app. I've do…

That's very interesting! My first thought, looking at the webpage: "Huh, that's neat. I didn't know that painting software didn't even attempt to do color mixing beyond naive inter…

Here's an idea: have the LLM output each comment with a "severity" score ranging from 0-100 or maybe a set of possible values ("trivial", "small", "high"). Let it get everything of…

That's certainly possible, but it reminds me a bit of a similar thing I've seen in their UI that rhymes in a way that makes me think otherwise. In the code interpreter tool, you ha…

For what it's worth, as a primarily backend dev having ~recently started getting more deeply into frontend web, I have specifically noted in my head that the box model isn't too in…

It does feel right to me, because it's not distilling the second model, and in fact the second model is not an image generation model at all, but a visual encoder. That is, it's a …

You may already be aware of it, but in case not - it sounds like tree-sitter-graph could be something you'd be interested in: https://docs.rs/tree-sitter-graph/latest/tree_sitter_g…

Or by the definition that the ratio between consecutive fib numbers approaches Phi, just multiply by 1.618? Though at that point might as well just use the real conversion ratio. I…

+1 on feeling there's a lot of UX possibilities left on the table. Most seem to have accepted chat as the only means of using LLMs. In particular, I don't think most people realize…

It does view RAW when compiled with the right flags. JXL too, interestingly. Managed to save a bunch of space on old photos (converting with cjxl, but which I wouldn't have done if…

Here's an idea: recursively mount code files/projects. Use something like tree-sitter to extract class and function definitions and make each into a "file" within the directory rep…

Have you tried asking it for a specific concrete length, like a number of words? I was also frustrated with concise answers when asking for long ones, but I found that the outputs …

My intuition is that a significant challenge for LLMs' ability to do arithmetics has to do with tokenization. For instance, `1654+73225` as per the OpenAI tokenizer tool breaks dow…

Interesting! Did you try the function calling API? I feel you with the line number troubles, it's hard to get something consistent there. Using diffs with GPT-4 isn't much better i…

I thought of something similar these days but with a different approach - rather than settrace, it would use a subclass of bdb.Bdb (the standard library base debugger, on top of wh…

How is suggesting the use of iterators and named tuples related to creating domain specific languages? If anything I'd say they're a much more generic and universally recognizable …

For all the brilliance in the AI and infra departments of OpenAI, their official Python library (which is the flagship one as I understand) feels pretty unidiomatic, designed witho…

Meta seems to actually be taking all the right steps in how they're contributing to open source AI research. Is this a "commodotize your complement" kind of situation?