Jeremy Howard @howard.fm · 11/05/2026No, the original MD release said: "Markdown allows you to write using an easy-to-read, easy-to-write plain text format, then convert it to structurally valid HTML." I.e originally designed for unformatted *writing*, but formatted *viewing*. For unformatted viewing .txt is appropriate. 130
Jeremy Howard @howard.fm · 01/05/2026I didn't YOLO it. txt is IMO the correct extension for a file designed to be viewed unformatted. The markdown is used for structure only. Web servers are widely set up to deliver .txt files OOTB. The aim is for something as consistent and compatible as possible. 160
Reposted by Jeremy Howard Martijn van den Broek @mfamyfuerza.bsky.social · 01/02/2026Great interview by @howard.fm on AI. Nice to see him continuing to mask. Jeremy Howard interview at PytorchCon with Anna Tong www.youtube.com/watch?v=LrFb...youtube.comJeremy Howard interview at PytorchCon with Anna TongYouTube video by Jeremy Howard 072
Reposted by Jeremy Howard Steve Pepple @stevepepple.bsky.social · 04/03/2026This is one of the absolute best conversations on AI and software I've heard in a while. (I listen to a lot of stuff about AI.) @howard.fm is an expert & practitioner who cuts through the noise & false dichotomies, to talk about the joys & pitfalls of human machine collab youtu.be/dHBEQ-Ryo24?...youtu.beThe Dangerous Illusion of AI Coding? - Jeremy HowardYouTube video by Machine Learning Street Talk 0314
Reposted by Jeremy Howard Pedro de las Heras Quirós @pheras.bsky.social · 19/03/20261/n "Getting better at the particular prompting skills or whatever details of the current generation AI CLI frameworks isn't growing. You know, that's as helpful as learning about the details of some AWS API when you don't actually understand how the internet works. 132
Jeremy Howard @howard.fm · 02/03/2026According to OpenAI, their contract with the US DoW locks in current law, "even if those laws or policies change in the future". Our legal analysis, with Virgil Law CEO Luke Versweyveld, shows that this is almost certainly incorrect. www.answer.ai/posts/2026-0...answer.aiCan a Contract Freeze the Law on Autonomous Weapons? – Answer.AIOpenAI claims its contract with the Department of War locks in current laws on autonomous weapons. Based on the contract language OpenAI has shared, 150 years of Supreme Court precedent, and the found... 1649
Reposted by Jeremy Howard Christian Laforte @chrlaf.bsky.social · 29/01/2026This is great news! Too few ML practitioners know what a gamechanger nbdev can be. Years ago at AMD I used nbdev (version 1?) to implement an early prototype of FSR4 realtime super resolution. It was self-explanatory (e.g. with diagrams showing examples of jittering and temporal super resolution)… 2144
Jeremy Howard @howard.fm · 29/01/2026Lots of folks have been asking for this for a while, but we wanted to wait until TOML support in the python stdlib was widely available. 3.11 added that support, and is pretty widely installed now. (We install `tomli` to backfill support where needed.) 050
Jeremy Howard @howard.fm · 29/01/2026nbdev v3 is released! The big change is that nbdev now uses pyproject.toml, not settings.ini, for all config. (nbdev v1 predates pyproject's support for project metadata, so we used our own settings file.) v3's `nbdev_migrate_config` automates migration nbdev.fast.ai/getting_star...nbdev.fast.aiGetting Started – nbdevCreate delightful software with Jupyter Notebooks 2455
Jeremy Howard @howard.fm · 23/01/2026Actually Claude Skills et al are just a re-implementation of this idea - i.e give LLMs access to links with descriptions to LLM-friendly info, and let it follow them if it wants to. 020
Reposted by Jeremy Howard Stefan Neumann @neumannstefan.com · 23/01/2026There is a third-party app for this, you can find it here: bsky-follow-finder.theo.io My impression with bluesky is that a lot of things exist but there is just no way of finding them. @nkgarg.bsky.socialbsky-follow-finder.theo.ioBluesky Network AnalyzerFind accounts that you don't follow (yet) but are followed by lots of accounts that you do follow. 3174
Reposted by Jeremy Howard Rachel Thomas @math-rachel.bsky.social · 21/01/2026Close reading is a technique for careful analysis of a piece of writing, practiced by many ancient cultures, major religions, & academic scholars. The latest fastai course experimented with using AI to go deeper when reading. 1/ www.fast.ai/posts/2026-0...fast.aiHow To Use AI for the Ancient Art of Close Reading – fast.aiExperiments in reading with LLMs 1194
Jeremy Howard @howard.fm · 20/01/2026llms.txt has nothing to do with SEO. It's basically a pre-written CLAUDE.md/AGENTS.md file to help people using stuff like Cursor and Claude Code use your site/product. It's not used for training LLMs. 160
Reposted by Jeremy Howard Eric Topol @erictopol.bsky.social · 16/01/2026GenAI has the potential to worsen health inequities. In a new review we show that with intent, confronting the challenges, it could be just the opposite, globally led by @nliulab.bsky.social nature.com/articles/s44...nature.comLarge language models in global health - Nature HealthLarge language models (LLMs) are emerging as powerful tools in healthcare, with a growing role in global health, particularly in low- and middle-income countries. This Perspective examines the current... 210937
Jeremy Howard @howard.fm · 15/01/2026I gave it 15. It found nothing of interest. I'm not gonna spend more time on this for now since it seems like a dead-end. 100
Jeremy Howard @howard.fm · 14/01/2026wdyt? (I haven't tried installing it, so I'm curious to know if I really should be avoiding it!) 110
Reposted by Jeremy Howard Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 12/01/2026Oh wow, deepseek is starting to make serious progress on LLMs that offload memory to external storage: github.com/deepseek-ai/...github.com 622026
Jeremy Howard @howard.fm · 12/01/2026(Sorry for the n00b questions, I hope they're not too annoying!) 020
Jeremy Howard @howard.fm · 12/01/2026Does clicking the "show less like this" button do anything on that feed BTW? Are there some docs or anything about how to get the best out of it? 430
Reposted by Jeremy Howard Sung Kim @sungkim.bsky.social · 12/01/2026ZLUDA, now in its third iteration, has added support for CUDA 13.1 compatibility on non-NVIDIA GPUs (well… AMD GPUs). - 1st iteration: Intel created ZLUDA as a drop-in replacement for CUDA on non-NVIDIA GPUs. - 2nd iteration: AMD took over development after Intel dropped support. 2223
Reposted by Jeremy Howard Scott McGrath @smcgrath.phd · 12/01/2026Anthropic announces Claude for Healthcare and expanded Life Sciences tools. By connecting directly to data sources like CMS and Medidata, Opus 4.5 aims to reduce administrative burdens. There are some promising ideas, but we need to see if they can deliver. #MedSky #MLSkyanthropic.comAdvancing Claude in healthcare and the life sciencesIntroducing Claude for Healthcare with HIPAA-ready infrastructure, plus expanded Life Sciences tools for clinical trials and regulatory submissions. New connectors to CMS, Medidata, and ClinicalTrials... 0303
Reposted by Jeremy Howard hardmaru @hardmaru.bsky.social · 12/01/2026One of my favorite findings: Positional embeddings are just training wheels. They help convergence but hurt long-context generalization. We found that if you simply delete them after pretraining and recalibrate for <1% of the original budget, you unlock massive context windows. Smarter, not harder. 821732
Reposted by Jeremy Howard Nathan Lambert @natolambert.bsky.social · 12/01/2026OpenAI's GPT OSS is still insanely underrated as a highly adopted open LLM. Downloads are out of control. 6573
Reposted by Jeremy Howard Tzu-Mao Li @tzumaoli.bsky.social · 09/01/2026cseweb.ucsd.edu/~tzli/novelt... I gave an internal talk at UCSD last year regarding "novelty" in computer science research. In it I "debunked" some of the myth people seem to have about what is good research in computer science these days. People seemed to like it, so I thought I should share.cseweb.ucsd.edu 27625
Reposted by Jeremy Howard Paul Graf @paulginva.bsky.social · 10/01/2026NEW: China is testing a megawatt-scale airborne wind system that captures stronger high-altitude winds, potentially producing up to 10× more energy with 40% less material and 30% lower costs. It could also be quickly deployed to supply emergency power after disasters. 731723477
Reposted by Jeremy Howard arXiv @arxiv.bsky.social · 01/01/2026Happy arxiv:2601 to all who celebrate! 😉🎉🔬 We are so thankful for everyone who uses arXiv everyday to share and discover new research. Happy new year to all - here's hoping for a 2026 full of #openaccess science & lots of new discoveries. 212619
Jeremy Howard @howard.fm · 11/01/2026Well I just opened bluesky and this was #1 on that list, which is absolutely content I'm here for :D 1200
Jeremy Howard @howard.fm · 11/01/2026I found that one already and subscribed, but it's not stuff I'm interested in. I need to find enough AI/ML posts to like to start getting it to notice, I guess... 160
Jeremy Howard @howard.fm · 11/01/2026I haven't been here for a year or so because the anti-AI bullying was too much. How does one get an AI/ML focused feed here nowadays and stay away from the hateful people? 15811
Reposted by Jeremy Howard Helen Toner @hlntnr.bsky.social · 05/04/2025Seems likely that at some point AI will make it much easier to hack critical infrastructure, create bioweapons, etc etc. Many argue that if so, a hardcore nonproliferation strategy is our only option. Rising Tide launch week post 3/3 is on why I disagree 🧵 helentoner.substack.com/p/nonprolife... 4174
Reposted by Jeremy Howard Dmytro Mishkin @ducha-aiki.bsky.social · 12/03/2025Explaining Human Preferences via Metrics for Structured 3D Reconstruction @jacklangerman.bsky.social Denys Rozumnyi, Yuzhong Huang, @ducha-aiki.bsky.social tl;dr: we asked 3D modelers to rank wireframe reconstructions & compared it to ranking by metrics. Observations🧵 1/ arxiv.org/abs/2503.08208 12810
Reposted by Jeremy Howard Carlo Lepelaars @carlolepelaars.bsky.social · 18/03/2025Struggling with Python `re`? regex is a superset of re with additional functionality and more thorough Unicode support. It is very powerful and still kind of an open secret. When I saw @howard.fm explain this I just had to write about it! Today I Learned: carlo.ai/til/regex-se...carlo.airegex.search - Carlo LepelaarsUsing regex.search for simple string parsing. `regex` is a superset of `re` with additional functionality and more thorough Unicode support. 1183
Reposted by Jeremy Howard Dr Noor Bari @njbbari3.bsky.social · 09/03/2025Currently reading a study about influenza aerosol transmission, published in 2020 and shows more RNA copies (log scale) in fine aerosol fraction. In 2020… and we still haven’t sorted out air safety in hospitals, and we are still whinging about flu being such a big problem (which it is). 819841