lsb

8 ms·

1.
by lsb 10d ago
The context window is limited to 64k or 128k. If you’re using this with a coding agent, there’s going to be a lot of compactions. I found that I had a subagent whose compacted context plus prompts and such was over the window and it errored
2.
by lsb 20d ago
As we have more powerful AI plus agents that can run on your phone, keeping everything as a pile of markdown plus some compute on top makes a lot of sense
3.
by lsb 22d ago
This is glitch art for text, I love it
4.
by lsb 26d ago
If this much JSON could fit in a microcontroller’s memory (7.5MB in text), and it’s a performance issue, maybe it’s worth upgrading the JSON parser?
5.
by lsb 28d ago
The page has comparisons with Qwen 3.6 27b and I’d love to see comparisons with Qwen 3.8 27b, the newer one is much more capable!
6.
by lsb 1mo ago
Google is making private AI practical with Gemma4 something that you can run without an Internet connection. All of the proofs of privacy rely on us getting the math right. All of the privacy from unplugging your internet cable is there by
7.
by lsb 1mo ago
Fixed!
8.
by lsb 1mo ago
Neat! Similar to my graduate thesis, NoDictionaries: https://nodictionaries.com/cato/de-agri-cultura/156
9.
by lsb 2mo ago
Homer is the Homer of America, unless you believe in fencing off universal properties of mankind into exclusive tribal ownership. (cf Saul Bellow asking “who is the Tolstoy of the Zulus”, and Ralph Wiley in the Atlantic saying that Tolstoy
10.
by lsb 4mo ago
Happy to answer any questions :)
11.
Show HN: Visualizing Tiny LLMs from OpenAI's Parameter Golf(leebutterman.com)
3 points by lsb 4mo ago|1 comments
12.
by lsb 5mo ago
Self Hosted object storage looks neat! For this project, where you have 120GB of customer data, and thirty requests a second for ~8k objects (0.25MB/s object reads), you’d seem to be able to 100x the throughput vertically scaling on on
13.
by lsb 6mo ago
The real world success they report reminds me of Simon Willison’s Red Green TDD: https://simonwillison.net/guides/agentic-engineering-pattern... > Instead of taking a stab in the dark, Leanstral rolled up its sleeve
14.
by lsb 7mo ago
It’s named after the multi-decade data compression test image https://en.wikipedia.org/wiki/Lenna Buy the book! https://qntm.org/vhitaos
15.
by lsb 8mo ago
I'm using Sonnet with 1M Context Window at work, just stuffing everything in a window (it works fine for now), and I'm hoping to investigate Recursive Language Models with DSPy when I'm using local models with Ollama
16.
by lsb 9mo ago
The New York Times has said that the US president has reported capturing the president of Venezuela https://www.nytimes.com/live/2026/01/03/world/trump-united-s... Source about aviation: primary (I
17.
Tell HN: No continental US flights due to attack on Venezuela
11 points by lsb 9mo ago|3 comments
18.
by lsb 9mo ago
This is super interesting! Apache Arrow is trying to do something similar, using Flatbuffer to serialize with zero-copy and zero-parse semantics, and an index structure built on top of that. Would love to see comparisons with Arrow
19.
by lsb 10mo ago
My threshold for “does not need to be smaller” is “can this run on a Raspberry Pi”. This is a helpful benchmark for maximum likely useful optimization. A Pi has 4 cores and 16GB of memory these days, so, running Qwen3 4B on a pi is pretty c
20.
by lsb 10mo ago
Happy to answer any questions you have :)
21.
Show HN: DSPy on a Pi: Cheap Prompt Optimization with GEPA and Qwen3(leebutterman.com)
4 points by lsb 10mo ago|1 comments
22.
by lsb 11mo ago
Curious about comparisons with Apache Arrow, which uses flatbuffers to avoid memory copying during deserialization, which is well supported by the Pandas ecosystem, and which allows users to serialize arrays as lists of numbers that have ha
23.
by lsb 1y ago
fast.ai (some of the authors of this) was transformative for me, and the community was super nice. Cannot recommend looking into this highly enough.
24.
by lsb 1y ago
This is halfbakery! I love it! (For example, a recent half baked idea there is a perpetually burning flag. https://www.halfbakery.com/idea/Perpetually_20Burning_20Flag... )
25.
by lsb 1y ago
How are you a landlord if you're paying property taxes? Once you have everything else set up, you can migrate to a server hosted on your own internet connection. Running your own data center is one of the more tricky parts of the equat
26.
by lsb 1y ago
Interesting! Text files in git can work for small sizes, like your 100MB. That is what's known in FAISS as a "flat" index, just one thing after another. And obviously you can query by primary key to the key-value store that i
27.
by lsb 1y ago
That’s wild that with a KV cache and compilation on the Mac CPU you are faster than on an A100 GPU.
28.
by lsb 1y ago
Also: this SSO tax is deceptively framed. Many of these services allow one to sign in through, for example, Google, which can count as a single sign on, and many organizations have a mail account, but that isn’t taken into account.
29.
by lsb 1y ago
This is evocative of “cramming”, a paper from a few years ago, where the author tried to find the best model they could train for a day on a modern laptop: https://arxiv.org/abs/2212.14034
30.
by lsb 1y ago
I used Claude Code to navigate a legacy codebase the other day, and having the ability to ask "how many of these files have helper methods that are duplicated or almost but not quite exactly duplicated?" was very much a superpow

More ›