LFM2.5-230M is a promising option for fine-tuning to translate natural language instructions into a series of tool calls. www.liquid.ai/blog/lfm2-5-...
LFM2.5-230M: Built to Run Anywhere — Blog liquid.ai / https://www.liquid.ai/blog/lfm2-5-230m Meet LFM2.5-230M, Liquid AI’s smallest model yet: a fast, open-weight foundation model for fine-tuning, edge deployment, tool use, and data extraction.Notes #
Short-form thoughts, saved links, and quoted fragments.
Drop-in replacement for whisper.cpp supporting STT with lots of different model families. I also appreciate that all the models were tested to compare WER. github.com/handy-comput...
https://github.com/handy-computer/transcribe.cpp github.com / https://github.com/handy-computer/transcribe.cppA two-stage privacy filter that runs completely on-device. First stage is a set of regex rules, second stage is a 14Mb MiniLM model fine tuned on the OpenPII dataset. This is where we need to be headed. ndstudio.gov/posts/say-he...
Introducing Rampart ndstudio.gov / https://ndstudio.gov/posts/say-hello-to-rampart An open-source, on-device personal information filter — among the smallest and most effective that can reasonably run in a browser.TIL - with pv (pipe viewer) you can add progress bars to any data transfer. docker save ubuntu:latest | pv -s $(docker images --format "{{.Size}}" ubuntu:latest) > ubuntu.tar
TIL - the Ettin model series exists, and they teach us: - Choose encoders for classification, retrieval, embeddings - Choose decoders for text generation - Size matters less than architecture choice (400M encoder beats 1B decoder on its specialized tasks) lighton.ai/lighton-blog...
Introducing Ettin Suite: the SoTA open recipe to outperform existing Generative & Retrieval Models - LightOn lighton.ai / https://lighton.ai/lighton-blogs/introducing-ettin-suite-the-sota-open-recipe-to-outperform-existing-generative-retrieval-models Introducing Ettin, the first ever SOTA suite of paired encoder & decoder models, developed by Johns Hopkins University in collaboration with LightOn.I really want this uv audit --fix github.com/astral-sh/uv...
Support a `--fix` type flag for `audit` to allow automatic fixing of vulnerabilities according to semantic versioning · Issue #19428 · astral-sh/uv github.com / https://github.com/astral-sh/uv/issues/19428 Summary I propose a --fix flag (or command) that would automatically fix vulnerabilities caught by the new audit command. Where I currently work, I wrote a script that uses a combination of the uv-...Some tips for safe package management with UV: 1. uv add pydantic --bounds major to ensure version is upper bound to avoid pulling in latest changes 2. Add a cooldown period in pyproject.toml to block latest versions [tool.uv] exclude-newer = "7 days" www.loopwerk.io/articles/202...
uv is fantastic, but its package management UX is a mess loopwerk.io / https://www.loopwerk.io/articles/2026/uv-ux-mess/ uv's CLI feels surprisingly clunky compared to its peers like pnpm or Poetry.Use this to get around stupid google AI
“Use this as your search engine query "https://www.google.com/search?q=%s&udm=14" :)”
Such a cool article about reward hacking in RL. goblins.mchen.workers.dev
how to train your goblin goblins.mchen.workers.dev / https://goblins.mchen.workers.dev/ what if other models could go goblin mode? an interactive walkthrough of training models with RL.Concerning. There have been other studies which produced the "performance" findings, but this is the first I've seen of the "persistence" issue.
“🚨New preprint and our results are rather concerning.. We find the "boiling frog" equivalent of AI use. Using large-scale RCTs, we provide *casual* evidence that AI assistance reduces persistence and hurts independent performance. And these effects emerge after just 10–15 minutes of AI use! 1/”
You can go one better and use fine-grained tokens. e.g. contents:read if all you need is read-only access to repo contents
“Btw, you can give gh cli a readonly token instead of the normal gh auth method which will have write. Drastically reduces blast radius if you are concerned about your gh cli token being stolen (which you should be) gh auth login —with-token hit enter Then paste the token, it will persist for you”
So many good ideas here, great resources
“What a day for AI resistance resources! 1) The AI Resist List from @karenhao.bsky.social and colleagues at DAIR at airesistlist.org 2) The Luddite Lab from @alexhanna.bsky.social and colleagues at DAIR at labor.dair-institute.org”
This is a step in the right direction but I don't think it gets us all the way. I see 3 distinct problems: 1. How do I see relevant updates/news from the sources/people I care about (maybe solved with this approach) 🧵
“Stop Begging Big Tech To Fix Your Social Media Experience. You Can Do It Yourself. Disclaimer: This post talks about Bluesky and an offering from Bluesky and I am on the Bluesky board. Take everything I say with whatever size grains of salt you feel is appropriate. I've written a few times now…”
substack.com/@polymathinv...
Polymath Investor (@polymathinvestor) substack.com / https://substack.com/@polymathinvestor/note/c-261699738?r=5yt43c A few months back, I published this guide on how to remember everything you read. Re-sharing it here for anyone who finds these protocols useful. (Part 1/2)I personally think this is a huge opportunity for competitors to position themselves as "human-centric" with a focus on delivering human generated and verified content on the internet. tante.cc/2026/05/20/o...
On Google declaring war on the Web tante.cc / https://tante.cc/2026/05/20/on-google-declaring-war-on-the-web/#:~:text=In%20Yesterday%E2%80%99s%20IO,links%20to%20information In Yesterday’s IO Keynote Google declared war on the remnants of the Web. (See longer description on their website.) TL;DR: They are pushing Search more into the “here’s your processed answer” directi...Are you re-raising exceptions just to add extra context? Since Python 3.11 you can attach notes to exceptions instead #PythonDev #python
#PythonDev #python
“pythonkoans.substack.com/p/koan-15-th...”
#PythonDev #python
“open.substack.com/pub/pythonko...”
open.substack.com/pub/pythonko...
Koan 20: The Unreliable Messenger open.substack.com / https://open.substack.com/pub/pythonkoans/p/koan-20-the-unreliable-messenger?utm_source=share&utm_medium=android&r=5yt43c Exploring traps in try/finally blocks, and the reasoning behind PEP601 and PEP765open.substack.com/pub/pythonko...
Koan 10: The Flowing River (Part 1) open.substack.com / https://open.substack.com/pub/pythonkoans/p/koan-10-the-flowing-river-part-1?utm_source=share&utm_medium=android&r=5yt43c Understanding how Python's list comprehensions work under the hoodVibecoding “floods the zone” for the developer. They get absolutely inundated with chat output as well as code diffs. No matter how much you “review” you’re never going to understand the code as much as if you had written it yourself.
One thing I detest about AI generated Python code is the ungodly amount of private helpers to do basic shit: def _strip_whitespace def _get_first_element Just stop.
pythonkoans.substack.com/p/koan-15-th...
Koan 15: The Invisible Ink pythonkoans.substack.com / https://pythonkoans.substack.com/p/koan-15-the-invisible-ink Understanding Unicode equivalence and the deceptive nature of glyphsUnderstanding Late Binding in Python Closures open.substack.com/pub/pythonko...
Koan 6: The Forgetful Calligrapher open.substack.com / https://open.substack.com/pub/pythonkoans/p/the-forgetful-calligrapher?r=5yt43c&utm_campaign=post&utm_medium=web Understanding Late Binding in Python Closures#python uses the flyweight design pattern to intern strings
Unravelling how Python finds and binds names, and why variables sometimes behave like shadows. pythonkoans.substack.com/p/koan-4-the...
Koan 4: The Grand Library pythonkoans.substack.com / https://pythonkoans.substack.com/p/koan-4-the-grand-library Understanding Python’s LEGB rule, closures, and how names echo through nested functions#python sets are useful when you need: - fast membership testing - deduplication or uniqueness guarantees - comparing two sets of items
#python list tips: - reversed(list) is more memory efficient for reversing a list than slicing list[::-1] because it returns an iterator instead of a copy - You can remove multiple items with slice assignment using list[1:4] = [] or del a[1:4]
#python variable scoping can be confusing if you’re used to other programming languages. A scope determines where a variable name is visible, and it’s lifetime. Python checks for the variable definition in the following order, and stops at the first match
It’s been a fun July writing Python Koans, I’m hoping to carry this momentum for the rest of the year! If you have suggestions or topics you want me to cover, feel free to DM me! #python pythonkoans.substack.com
Python Koans | Vivis Dev | Substack pythonkoans.substack.com / https://pythonkoans.substack.com Python lessons wrapped in koans. Small puzzles, deep truths. Not your usual tutorial thread. Click to read Python Koans, by Vivis Dev, a Substack publication with hundreds of subscribers.ChatGPT has started quietly nudging people towards their search extension. It's been out since December 2024, and has 4M users. Will this make a dent into Google?