AI 圈的“信息源头“,就这 30 个人。 These 30 accounts power the global AI signal. 刷 100 个二手解读,不如关注 10 个一手账号。 Skip the 100 hot takes. Follow the 10 people they’re quoting. 这份清单覆盖 7 个维度: This list covers 7 dimensions: 研究 · 工程 · 教育 · 开源 · 评测 · AGI · 全球生态 Research · Engineering · Education · Open Source · Evals · AGI · Global Ecosystem 按这个清单刷一周,认知密度会有质变 Follow them for a week. Your AI literacy will compound ━━━━━━━━━ 研究 · 顶级心智 / Research · Top Minds ━━━━━━━━━ ▍@karpathy 推文常常提前定义 LLM 叙事。 Often defines the LLM narrative months before everyone else. ▍@fchollet Keras 作者,ARC-AGI 提出者。关于智能本质的深度思考。 Creator of Keras and ARC-AGI. Deep thinking on what intelligence really is. ▍@ylecun 深度学习先驱,Meta 首席 AI 科学家。常对主流路线做批判性讨论。 Deep learning pioneer, Meta’s Chief AI Scientist. Sharp critiques of mainstream paths. ▍@lilianweng 前 OpenAI 成员,Lil’Log 作者。LLM 技术细节拆解的天花板。 Former OpenAI, author of Lil’Log. The gold standard for technical deep dives. ▍@drfeifei 计算机视觉先驱。关注以人为中心的 AI 与空间智能。 CV pioneer. Focused on human-centered AI and spatial intelligence. ▍@demishassabis Google DeepMind CEO。理解 AGI 路线最重要的窗口。 CEO of Google DeepMind. The clearest window into the AGI roadmap. ▍@sama OpenAI CEO。发言常是行业风向标。 OpenAI CEO. His posts often move the entire industry. ━━━━━━━━━ 工程 · 实操与系统 / Engineering · Build & Ship ━━━━━━━━━ ▍@simonw Django 联合创始人。LLM 工具、Agent、Prompt 工程的实操王。 Co-creator of Django. The go-to source for hands-on LLM tooling and agents. ▍@rasbt Sebastian Raschka。“从零构建” 系列教程的标杆。 The benchmark for “build it from scratch” tutorials. ▍@swyx Latent Space 主理人。“AI Engineer” 概念的推广者。 Host of Latent Space. The one who put “AI Engineer” on the map. ▍@ggerganov llama.cpp 作者。本地推理、量化、端侧 AI 绕不开的人。 Creator of llama.cpp. You can’t talk local inference without him. ▍@hwchase17 LangChain 创始人。Agent 框架演化的第一手信息源。 Founder of LangChain. First-hand source on agent framework evolution. ▍@amasad Replit CEO。AI Coding 产品化最敏锐的观察者之一。 CEO of Replit. One of the sharpest voices on AI-native developer tools. ━━━━━━━━━ 教育 · 入门到进阶 / Education · From Zero to Deep ━━━━━━━━━ ▍@AndrewYNg AI 教育领域的传奇。ML 建议、课程、产品落地全覆盖。 A legend in AI education. ML advice, courses, and real-world deployment. ▍@jeremyphoward http://fast.ai 创始人。长期推动实用深度学习的普及。 Founder of http://fast.ai. Has spent years democratizing deep learning. ▍@maximelabonne LLM 微调实战派。从书到代码全开源。 Fine-tuning in practice. Books, code, all open-sourced. ━━━━━━━━━ 开源 · 社区脉搏 / Open Source · Community Pulse ━━━━━━━━━ ▍@_akhaliq arXiv 论文、模型发布、开源项目的信息流,更新极快。 A live feed of arXiv papers, model drops, and open-source projects. ▍@dair_ai 高频 AI 论文线程,通俗拆解前沿研究。 Frequent paper threads that make frontier research digestible. ▍@Teknium1 Nous Research 联合创始人。开源 LLM 训练与微调的代表人物。 Co-founder of Nous Research. A leading voice in open-source LLM training. ▍@ClementDelangue Hugging Face CEO。开源 AI 生态最高质量的信息源之一。 CEO of Hugging Face. One of the highest-quality signals in open-source AI. ━━━━━━━━━ 评测 · 真实能力 / Evals · What Models Actually Do ━━━━━━━━━ ▍@goodside LLM 评测、Prompt 研究、真实能力测试。模型行为观察非常细。 LLM evals, prompt research, real capability probes. Razor-sharp observations. ▍@lmsysorg Chatbot Arena 官方账号。模型真实对战最权威的赛场。 The official Chatbot Arena. The most credible head-to-head model leaderboard. ▍@scaling01 Scaling Laws、模型对比、独立 Benchmark。数据派必看。 Scaling laws, model comparisons, independent benchmarks. A must for data nerds. ━━━━━━━━━ AGI · 长期主义 / AGI · The Long Game ━━━━━━━━━ ▍@ID_AA_Carmack 关注 AGI 与底层优化。重新思考“智能“与“工程“。 On AGI and low-level optimization. Will rewire how you think about intelligence. ▍@gwern 高质量长文作者。AI 研究笔记与长期主义视角。 Long-form essays. AI research notes through a long-term lens. ▍@jackclarkSF Anthropic 联合创始人,Import AI 周报作者。AI 政策与产业结构的高密度观察。 Co-founder of Anthropic, author of Import AI. Dense takes on policy and industry shifts. ▍@AnthropicAI Anthropic 官方账号。Claude 团队的研究节奏与安全工作。 The official Anthropic account. Research cadence and safety work from the Claude team. ▍@ESYudkowsky AI Alignment 领域最具影响力的声音之一。立场强烈,但绕不开。 One of the most influential voices in alignment. Strong opinions, unavoidable reading. ━━━━━━━━━ 中国 & 全球生态 / China & Global Ecosystem ━━━━━━━━━ ▍@teortaxesTex 持续追踪 DeepSeek / Qwen 等中国 AI 动态。 The best non-Chinese tracker of DeepSeek, Qwen, and the China AI scene. ▍@deepseek_ai 国产开源大模型代表。模型发布与技术报告第一手来源。 A flagship of Chinese open-source LLMs. First-hand model and research drops. ━━━━━━━━━ 15 → 30,这是一份“信息源头“清单。 15 → 30. This is a list of sources, not influencers. 不是热门,不是 KOL,是真正在生产 AI 一手认知的人。 Not hype accounts. The people actually producing the signal. 收藏 + 转发,然后挨个关注 Bookmark, repost, then follow them all 还有谁该加?评论区补 Who else should be on this list? Drop them below

fast.ai - fast.ai—Making neural nets uncool again

Source: https://www.fast.ai/

### I Don’t Want a Learning Dashboard for My Childeducation

technical

What analog and AI education both get wrong Feb 17, 2026 Rachel Thomas

### Breaking the Spell of Vibe Codingai-in-society

technical

Sinister variations on the positive state of flow Jan 28, 2026 Rachel Thomas

### How To Use AI for the Ancient Art of Close Readingeducation

technical

Experiments in reading with LLMs Jan 21, 2026 Rachel Thomas

### Stop Saying Boredom is Good for Kidseducation

Chronic boredom causes stress, disengagement, and poor well-being in adults. So why do we glorify it for children? Dec 3, 2025 Rachel Thomas

### A Guide to Solveit Featurestechnical

An overview of the features of the Solveit platform, which is designed to make exploration and iterative development easier and faster. Nov 7, 2025 Solveit and Kerem Turgutlu

### Build to Lastadvice

technical

Chris Lattner on software craftsmanship and AI Oct 30, 2025 Jeremy Howard

### Let’s Build the GPT Tokenizer: A Complete Guide to Tokenization in LLMstechnical

A text and code version of Karpathy’s famous tokenizer video. Oct 16, 2025 Andrej Karpathy, via Solveit and Kerem Turgutlu

### How to Solve it With Code course now availablecourses

An email sent to all fast.ai forum users. Oct 15, 2025 Jeremy Howard

### fasttransform: Reversible Pipelines Made Simpletechnical

Introducing fasttransform, a Python library that makes data transformations reversible and extensible through the power of multiple dispatch. Feb 20, 2025 Rens Dimmendaal, Hamel Husain, & Jeremy Howard

### What AI can tell us about microscope slideshealth

technical

A friendly introduction to Foundation Models for Computational Pathology Feb 3, 2025 Rachel Thomas

### A New Chapter for fast.ai: How To Solve It With Codeeducation

courses

fast.ai is joining Answer.AI, and we’re announcing a new kind of educational experience, ‘How To Solve It With Code’ Nov 7, 2024 Jeremy Howard

### In defense of screen timeeducation

Pundits say my husband and I are parenting wrong. Oct 29, 2024 Rachel Thomas

### A new old kind of R&D labai-in-society

technical

Answer.AI is a new kind of AI R&D lab which creates practical end-user products based on foundational research breakthroughs. Dec 12, 2023 Jeremy Howard

### Can LLMs learn from a single example?technical

We’ve noticed an unusual training pattern in fine-tuning LLMs. At first we thought it’s a bug, but now we think it shows LLMs can learn effectively from a single example. Sep 4, 2023 Jeremy Howard and Jonathan Whitaker

### The real risk of AI is how it concentrates powerai-in-society

Moving AI ethics beyond explainability and fairness to justice Jul 29, 2023 Rachel Thomas

### AI Safety and the Age of Dislightenmentai-in-society

Model licensing & surveillance will likely be counterproductive by concentrating power in unsustainable ways Jul 10, 2023 Jeremy Howard

### Is Avoiding Extinction from AI Really an Urgent Priority?ai-in-society

The history of technology suggests that the greatest risks come not from the tech, but from the people who control it May 30, 2023 Seth Lazar, Jeremy Howard, & Arvind Narayanan

### Mojo may be the biggest programming language advance in decadestechnical

Mojo is a new programming language, based on Python, which fixes Python’s performance and deployment problems. May 4, 2023 Jeremy Howard

### From Deep Learning Foundations to Stable Diffusioncourses

We’ve released our new course with over 30 hours of video content. Apr 4, 2023 Jeremy Howard

### GPT 4 and the Uncharted Territories of Languageai-in-society

Language is a source of limitation and liberation. GPT 4 pushes this idea to the extreme by giving us access to unlimited language. Mar 20, 2023 Jeremy Howard

### I was an AI researcher. Now, I am an immunology student.education

Last year, I became captivated by a new topic in a way that I hadn’t felt since I first discovered machine learning Feb 7, 2023 Rachel Thomas

### 1st Two Lessons of From Deep Learning Foundations to Stable Diffusioncourses

4 videos from Practical Deep Learning for Coders Part 2, 2022 have been released as a special early preview of the new course. Oct 19, 2022 Jeremy Howard

### Deep Learning Foundations Signup, Open Source Scholarships, & Morecourses

Signups are now open for Practical Deep Learning for Coders Part 2, 2022. Scholarships are available for fast.ai community contributors, open source developers, and diversity scholars. Sep 23, 2022 Jeremy Howard

### From Deep Learning Foundations to Stable Diffusioncourses

Practical Deep Learning for Coders part 2, 2022 Sep 16, 2022 Jeremy Howard

### My family’s unlikely homeschooling journeyeducation

Prior to 2020, we never expected to homeschool, and now we have committed to it long-term. Sep 6, 2022 Rachel Thomas

### The Jupyter+git problem is now solvedtechnical

Previously, using git with Jupyter could create conflicts and break notebooks. With nbdev2, the problem has been totally solved. Aug 25, 2022 Jeremy Howard

### nbdev+Quarto: A new secret weapon for productivitytechnical

Our favorite tool for software engineering productivity–nbdev, now re-written with Quarto Jul 28, 2022 Hamel Husain and Jeremy Howard

### Practical Deep Learning for Coders 2022courses

A complete from-scratch rewrite of fast.ai’s most popular course, that’s been 2 years in the making. Jul 21, 2022 Jeremy Howard

### Masks for COVID: Updating the evidencehealth

Notes taken whilst preparing a paper on mask efficacy from Nov to Jan 2022. My previous paper on this was written in April 2020 and published in the Proceedings of the National Academy of Science Jul 4, 2022 Jeremy Howard

### Qualitative humanities research is crucial to AIai-in-society

Following the thread of any seemingly quantitative issue in AI ethics (such as determining if software to rate loan applicants is racially biased or evaluating YouTube’s recommendation system) quickly leads to a host of qualitative questions. Unfortunately, there is often a large divide between computer scientists and social scientists, with over-simplified assumptions and fundamental misunderstandings of one another. Jun 1, 2022 Louisa Bartolo and Rachel Thomas

### AI Harms are Societal, Not Just Individualai-in-society

In the west, our ideas of harm are largely anchored to an individual being harmed by a particular action at a discrete moment in time. Yet the harms caused by algorithmic systems are often collective and communal. May 17, 2022 Rachel Thomas and Louisa Bartolo

### There’s no such thing as not a math personadvice

education

Many cultural factors, misconceptions, stereotypes, and obstacles turn people off to math. Mar 15, 2022 Rachel Thomas

### 7 Great Lightning Talks Related to Data Science Ethicsai-in-society

Lightning talks by Australian experts on a range of topics related to data science ethics, including machine learning in medicine, explainability, Indigenous-led AI, and the role of policy Mar 14, 2022 Rachel Thomas

### Doing Data Science for Social Good, Responsiblyai-in-society

While data for social good projects can be useful, there are also pitfalls to avoid. Nov 23, 2021 Rachel Thomas

### Avoiding Data Disasterstechnical

ai-in-society

Things can go disastrously wrong on data science and machine learning projects when we undervalue data work, use data in contexts that it wasn’t gathered for, or ignore the crucial role that humans play in the data science pipeline. Nov 4, 2021 Rachel Thomas

### SARS-CoV-2 Spike Protein Impairment of Endothelial Function Does Not Impact Vaccine Safetyhealth

All approved SARS-CoV-2 vaccines provide far more benefits than risks. The very rare risk of VITT from the AZ and JJ vaccines is not due to the spike proteins, but is most likely due to details of their formulation. Oct 27, 2021 Jeremy Howard and Uri Manor

### Statistical problems found when studying Long Covid in kidshealth

technical

Statistical tests need to be paired with proper data and study design to yield valid results. A recent review paper on Long Covid in children provides a useful example of how researchers can get this wrong. We use causal diagrams to decompose the problem and illustrate where errors were made. Oct 17, 2021 Jeremy Howard

### Medicine is Politicalhealth

From covid-19 to HIV research to the long history of wrongly assuming women’s illnesses are psychosomatic, we have seen again and again that medicine, like all science, is political. Oct 12, 2021 Rachel Thomas

https://www.fast.ai/posts/2021-09-25-new-yorker.html

### Inaccuracies, irresponsible coverage, and conflicts of interest in The New Yorkerhealth

Letter to the editor of the New Yorker on the irresponsible description of a suicide, undisclosed financial conflicts of interest, and omission of relevant medical research and historical context in their recent long covid article Sep 25, 2021 Rachel Thomas

### Australia can, and must, get R under 1.0health

By using better masks, monitoring and improving indoor air quality, and rolling out rapid tests, we could quickly halt the current outbreaks in the Australian states of New South Wales (NSW) and Victoria. If we fail to do so, and open up before 80% of all Australians are vaccinated, we may have tens of thousands of deaths, and hundreds of thousands of children with chronic illness which could last for years. Sep 3, 2021 Jeremy Howard

### 11 Short Videos About AI Ethicsai-in-society

I made a playlist of 11 short videos (most are 6-13 mins long) on Ethics in Machine Learning. This is from my ethics lecture in Practical Deep Learning for Coders v4. I thought these short videos would be easier to watch, share, or… Aug 16, 2021 Rachel Thomas

### Getting Specific about AI Risks (an AI Taxonomy)ai-in-society

The term “Artificial Intelligence” is a broad umbrella, referring to a variety of techniques applied to a range of tasks. This breadth can breed confusion. Success in using… Aug 16, 2021 Rachel Thomas

https://www.fast.ai/posts/2020-02-13-fastai-A-Layered-API-for-Deep-Learning.html

### fastai A Layered API for Deep Learning **Abstract:**fastai is a deep learning library which provides practitioners with high-level components that can quickly and easily provide state-of-the-art results in standard deep learning… Aug 13, 2021 Jeremy Howard and Sylvain Gugger

### fastdownload: the magic behind one of the famous 4 lines of codetechnical

Summary:Today we’re launching fastdownload, a new library that makes it easy for your users to download, verify, and extract archives. Aug 2, 2021 Jeremy Howard

### Is GitHub Copilot a blessing, or a curse?technical

GitHub Copilot is a new service from GitHub and OpenAI, described as “Your AI pair programmer”. It is a plugin to Visual Studio Code which auto-generates code for you based on the contents of the… Jul 19, 2021 Jeremy Howard

### fastchan, a new conda mini-distributiontechnical

Summary: today we’re announcing fastchan, a new conda mini-distribution with a focus on the PyTorch ecosystem. Using fastchan, installation and updates of libraries such as PyTo… Jul 15, 2021 Jeremy Howard

### 20 Years of Tech Startup Experiences in One Hourai-in-society

I’ve just returned to Australia to live, after a decade as an entrepreneur in San Francisco. For my first in-person talk in Australia, I shared my thoughts on how to build a… May 26, 2021 Jeremy Howard

### I violated a code of conductai-in-society

Update Oct 30, 2020: NumFOCUS has apologized to me. I accept their apology. I do not accept their assertion that “At the time of the interview, the committee had not determined that there was a violation of the code… Oct 28, 2020 Jeremy Howard

### Avoiding the smoke - how to breath clean airadvice

health

You can use filters in your mask, your home central air, and your fans to clean the air in your home. Sep 11, 2020 Jeremy Howard

### fast.ai releases new deep learning course, four libraries, and 600-page bookcourses

technical

We’re releasing Practical Deep Learning for Coders (2020), fastai v2, fastcore, and fastgpu. Aug 21, 2020 Jeremy Howard

### Forward from the ‘Deep Learning for Coders’ Bookcourses

To celebrate the release of fast.ai’s new course, book, and software libraries on August 21st, we’re making available the foreword that Soumith Chintala (the co-creator of PyTorch) wrote for the book Aug 20, 2020 Soumith Chintala

### Applied Data Ethics, a new free course, is essential for all working in techai-in-society

courses

Free, online course from fast.ai and University of San Francisco Data Institute covering disinformation, bias & fairness, ethical foundations, practical tools, privacy & surveillance, the silicon valley ecosystem, and algorithmic colonialism. Aug 19, 2020 Rachel Thomas

### Essential Work-From-Home Advice: Cheap and Easy Ergonomic Setupsadvice

health

You can permanently damage your back, neck, and wrists from working without an ergonomic setup. Learn how to create one for less at home. Aug 6, 2020 Rachel Thomas

### Cloth masks can protect the wearerhealth

DIY masks aren’t just for protecting those around you. They can protect you too, but material and fit matters a lot. Learn what works, based on the latest research. Jul 10, 2020 Jeremy Howard

### Particle sizes for mask filtrationhealth

SARS-CoV-2 does not float in the air. It’s expelled as large droplets, which are easily caught by a cloth mask. Jun 26, 2020 Jeremy Howard

### Introducing the first cohort of USF CADE Data Ethics Research Fellowsai-in-society

The University of San Francisco Center for Applied Data Ethics welcomes its first cohort of research fellows. Jun 16, 2020 Rachel Thomas

### Masks - FAQ for Skepticshealth

Wearing a mask decreases the number of people infected by an infectious wearer, because it reduces by around 99% the number of droplets that are ejected during speech Apr 20, 2020 Jeremy Howard

### Masks for all? The science says yes.health

Most scientific evidence points in the same direction: keep your droplets to yourself - wear a mask Apr 13, 2020 Professor Trisha Greenhalgh OBE and Jeremy Howard

### 6 Important Videos about Tech, Ethics, Policy, and Governmentai-in-society

At the CADE Tech Policy Workshop, members of local government, activists, researchers, and those in industry shared prespectives on how to understand the uses, risks, and opporutnities around government use of technology. Mar 31, 2020 Rachel Thomas

### Saving The Maskhealth

An initiative from STEM teachers in Hong Kong to make reusable masks, for the community, by the community Mar 20, 2020 Sarada Lee

### Covid-19, your community, and you - a data science perspectivehealth

We are data scientists—that is, our job is to understand how to analyze and interpret data. When we analyze the data around… Mar 9, 2020 Jeremy Howard and Rachel Thomas

### Disinformation: what it is, why it’s pervasive, and proposed regulationsai-in-society

At the CADE Tech Policy Workshop, experts Renee DiResta and Guillaume Chaslot spoke on disinformation, including the dyanmics that cause it to go viral and attempts towards addressing it. Feb 26, 2020 Rachel Thomas

### Tech Ethics Crisis: The Big Picture, and How We Got Hereai-in-society

At the CADE Tech Policy Workshop, Y-Vonne Hutchinson spoke about the role of tech in facilitating mass atrocity, and Catherine Bracy spoke on the need for empathy and collective action. Feb 9, 2020 Rachel Thomas

### 4 Principles for Responsible Government Use of Technologyai-in-society

As governments consider new uses of technology, in public places, this raises issues around surveillance of vulnerable populations, unintended consequences, and potential misuse. There are several principles to keep in mind in how these decisions can be made in a healthier and more responsible manner. Jan 21, 2020 Rachel Thomas

### Blogging with Jupyter Notebookstechnical

With Jupyter Notebooks and fast_template, we can easily share prose, code, tables, charts, and more! Jan 20, 2020 Jeremy Howard

### Your own blog with GitHub Pages and fast_template (4 part tutorial)technical

Overview of 4 part series explaining how to host your own blog without any coding Jan 20, 2020 Jeremy Howard

### Blogging with screenshotstechnical

We show how to use fast_template’s special ‘screenshot’ feature to get high-resolution screenshots Jan 19, 2020 Jeremy Howard

### Syncing your blog with your PC, and using your word processortechnical

With GitHub Pages you can synchronize your blog with your computer, and write posts with MS Word or Google Docs Jan 18, 2020 Jeremy Howard

### Your own hosted blog, the easy, free, open way (even if you’re not a computer expert)technical

An easy and free approach to using an entirely browser-based interface for all your blogging needs Jan 16, 2020 Jeremy Howard

### Self-supervised learning and computer visiontechnical

Pretrain a model using labels that are naturally part of the input data, instead of using external labels Jan 13, 2020 Jeremy Howard

https://www.fast.ai/posts/2020-01-07-data-questionnaire.html

### Data project checklisttechnical

There’s a lot more to creating useful data projects than just training an accurate model Jan 7, 2020 Jeremy Howard

https://www.fast.ai/posts/2019-11-27-nbdev.html

### nbdev: use Jupyter Notebooks for everythingtechnical

nbdev is a Python programming environment which allows you to create complete python packages, including tests and a rich documentation system, all in Jupyter Notebooks Dec 2, 2019 Jeremy Howard

### Concerned about the impacts of data misuse? Ways to get involved with the USF Center for Applied Data Ethicsai-in-society

Three months after the launch of CADE, find out what we’ve been up to and how you can get involved. Nov 4, 2019 Rachel Thomas

### The problem with metrics is a big problem for AIai-in-society

technical

Unthinkingly optimizing metrics can lead to a variety of grave harms, and what most current AI approaches do is to optimize metrics. Sep 24, 2019 Rachel Thomas

### 8 Things You Need to Know about Surveillanceai-in-society

8 important truths about surveillance Aug 7, 2019 Rachel Thomas

### Make Delegation Work in Python Fixing the delegation program usingdelegatesdecorator andGetAttr Aug 6, 2019 Jeremy Howard

### USF Launches New Center of Applied Data Ethicsai-in-society

courses

University of San Francisco is launching a Center of Applied Data Ethics, and Rachel Thomas is the director of the new center. Aug 5, 2019 Rachel Thomas

### new fast.ai course: A Code-First Introduction to Natural Language Processingtechnical

courses

fast.ai’s newest course is Code-First Intro to NLP. It covers a blend of traditional NLP techniques, recent deep learning approaches, and urgent ethical issues. Jul 8, 2019 Rachel Thomas

### Deep Learning from the Foundationscourses

Today we are releasing a new course, Deep Learning from the Foundations, which shows how to build a state of the art deep learning model from scratch. Jun 28, 2019 Jeremy Howard

### A LaTeX add-in for PowerPoint - my father’s day projecttechnical

For creating presentations there’s a lot of features in PowerPoint that are hard to beat. So it’s not surprising that it’s a very popular tool—I see a lot of folks… Jun 17, 2019 Jeremy Howard

### Was this Google Executive deeply misinformed or lying in the New York Times?ai-in-society

YouTube has played a significant role in radicalizing people into conspiracy theories that promote white supremacy, anti-vaxxing, denial of mass shootings, c… May 28, 2019 Rachel Thomas

### Advice for Better Blog Postsadvice

A blog is like a resume, only… May 13, 2019 Rachel Thomas

### Decrappification, DeOldification, and Super Resolutioncourses

We presented this work at the Facebook f8 conference. You can see this video of our talk here, or read on for more details and examples. May 3, 2019 Jason Antic (Deoldify), Jeremy Howard (fast.ai), and Uri Manor (Salk Institute)

### 16 Things You Can Do to Make Tech More Ethical, part 3ai-in-society

This post is part 3 in a series. Please check out part 1 here and part 2 here. May 3, 2019 Rachel Thomas

### 16 Things You Can Do to Make Tech More Ethical, part 2ai-in-society

This post is part 2 in a 3-part series. Please check out part 1 and part 3 as well. Apr 25, 2019 Rachel Thomas

### 16 Things You Can Do to Make Tech More Ethical, part 1ai-in-society

This post is part 1 in a 3-part series. Please check out part 2 and part 3 as well. Apr 22, 2019 Rachel Thomas

### fast.ai Embracing Swift for Deep Learningtechnical

Note from Jeremy: If you want to… Mar 6, 2019 Jeremy Howard

### A Conversation about Tech Ethics with the New York Times Chief Data Scientistai-in-society

Note from Rachel: Although I’m excited about the positive potential of tech, I’m also scared about the ways that tech is having a negative impact on society, and I’m interested in how we can push tech companies to do… Mar 4, 2019 Rachel Thomas and Chris Wiggins

### Dairy farming, solar panels, and diagnosing Parkinson’s disease: what can you do with deep learning?ai-in-society

courses

Many people incorrectly assume that AI is only for an elite few– a handful of Silicon Valley computer science prodigies with monthly budgets larger than most people’s… Feb 21, 2019 Rachel Thomas

### Some thoughts on zero-day threats in AI, and OpenAI’s GPT-2ai-in-society

There’s been a lot of discussion in the last couple of days about OpenAI’s new language model. OpenAI made the unusual decision to not release their trained model (the AI community is usually extremely open about sharing them). On the whole, the reaction has been… Feb 15, 2019 Jeremy Howard

### fastec2: AWS computer management for regular folkstechnical

This is part 1 of a series on fastec2. To learn how to run and monitor long-running tasks with fastec2 check out part 2. Feb 15, 2019 Jeremy Howard

### fastec2 script: Running and monitoring long-running taskstechnical

This is part 2 of a series on fastec2. For an introduction to fastec2, see part 1. Feb 15, 2019 Jeremy Howard

### Five Things That Scare Me About AIai-in-society

AI is being increasingly used to make important decisions. Many AI experts (including Jeff Dean, head of AI at Google, and Andrew Ng, founder of Coursera and deeplearning.ai) say that warnings about sentient robots are overblown, but other harms are not getting enough… Jan 29, 2019 Rachel Thomas

### fast.ai Diversity Fellows and Sponsors Wantedcourses

This post was originally published on 2018-08-16, but has been updated for the newest, upcoming course. Jan 28, 2019 Rachel Thomas

### Practical Deep Learning for Coders 2019courses

Launching today, the 2019 edition of Practical Deep Learning for Coders, the third iteration of the course, is 100% new material, including applications that have never been covered by an introductory deep learning course before (with some… Jan 24, 2019 Jeremy Howard

### C++11, random distributions, and Swifttechnical

Generating numbers from random distributions is a practically useful tool that any coder is likely to need at some point. C++11 added a rich set of random distribution… Jan 13, 2019 Jeremy Howard

### High Performance Numeric Programming with Swift: Explorations and Reflectionstechnical

Over the past few weeks I’ve been working on building some numeric programming libraries for Swift. But wait, isn’t Swift just what iOS programmers use for building… Jan 10, 2019 Jeremy Howard

### One year of deep learningadvice

My resolution for 2018 was to get into deep learning. I had stumbled upon a website called fast.ai in October 2017 after reading an article from the New York Times… Jan 2, 2019 Sylvain Gugger

### The new fast.ai research datasets collection, on AWS Open Datacourses

technical

In machine learning and deep learning we can’t do anything without data. So the people that create datasets for us to train our models are the (often under-appreciated)… Oct 16, 2018 Jeremy Howard (fast.ai) and Jed Sundwall (Open Data Global Lead, AWS)

### fastai v1 for PyTorch: Fast and accurate neural nets using modern best practicestechnical

Note from Jeremy: Want to learn more? Listen to me discuss fastai with Sam Charrington in this in-depth interview. Oct 2, 2018 Jeremy Howard

https://www.fast.ai/posts/2018-09-26-ml-launch.html

### Introduction to Machine Learning for Coders: Launchcourses

Today we’re launching our newest (and biggest!) course, Introduction to Machine Learning for Coders. The course, recorded at the University of San Francisco as part of the Masters of Science in Data… Sep 26, 2018 Jeremy Howard

### AI Ethics Resourcesadvice

ai-in-society

My newest Ask-A-Data-Scientist post was inspired by a computer science student who wrote in asking for advice on how to pursue a… Sep 24, 2018 Rachel Thomas

### What You Need to Know Before Considering a PhDadvice

education

My newest Ask-A-Data-Scientist post addresses the question of whether to pursue a PhD. You can find my previous Ask-A-Data-Scientist advice columns here. Aug 27, 2018 Rachel Thomas

### Practical Deep Learning for Coders, part-time Diversity Fellowships, Fall 2018courses

At fast.ai, we want to do… Aug 16, 2018 Rachel Thomas

### Now anyone can train Imagenet in 18 minutestechnical

This post extends the work described in a previous post, Training Imagenet in 3 hours for $25; and CIFAR10 for $0.26. Aug 10, 2018 Jeremy Howard

### What HBR Gets Wrong About Algorithms and Biasai-in-society

The Harvard Business Review recently published an article, Want Less-Biased Decisions? Use Algorithms. by Alex P. Miller. The article focuses on the fact that humans make very biased… Aug 7, 2018 Rachel Thomas

### Google’s AutoML: Cutting Through the Hypetechnical

This is part 3 in a series. Part 1 is here and Part 2 is here. Jul 23, 2018 Rachel Thomas

### An Opinionated Introduction to AutoML and Neural Architecture Searchtechnical

This is part 2 in a series. Check out part 1 here and part 3 here. Jul 16, 2018 Rachel Thomas

### What do machine learning practitioners actually do?technical

This post is part 1 of a series. Part 2 is an opinionated introduction to AutoML and neural architecture search, and Part 3 looks at Google’s AutoML in particular. Jul 12, 2018 Rachel Thomas

### AdamW and Super-convergence is now the fastest way to train neural netstechnical

*Note from Jeremy:*Welcome to fast.ai’s first scholar-in-residence, Sylvain Gugger. What better way to introduce him than to publish the results of his first research project at fast.ai. We’ll be using the results of this research to change how we train… Jul 2, 2018 Sylvain Gugger and Jeremy Howard

### Launching Cutting Edge Deep Learning for Coders: 2018 editioncourses

Today we are launching the 2018 edition ofCutting Edge Deep Learning for Coders, part 2 of fast.ai’sfreedeep learning course. Just as with our part 1Practical… May 7, 2018 Jeremy Howard

### Training Imagenet in 3 hours for USD 25; and CIFAR10 for USD 0.26technical

Posted: May 2, 2018 Apr 30, 2018 Jeremy Howard

### An Introduction to Deep Learning for Tabular Datatechnical

There is a powerful technique that is winning Kaggle competitions and is widely used at Google (according to Jeff Dean), Pinterest, and Instacart, yet that many people… Apr 29, 2018 Rachel Thomas

### What you need to know about Facebook and Ethicsai-in-society

While many news outlets have covered what Zuckerberg wore for his testimony before congress last week, I wish that several more substantial issues were getting greater coverage, such as the lackluster response of Facebook to its… Apr 19, 2018 Rachel Thomas

### A Discussion about Accessibility in AI at Stanfordadvice

I recently was a guest speaker at the Stanford AI Salon on the topic of accessiblity in AI, which included a… Apr 10, 2018 Rachel Thomas

### Adding Data Science to your College Curriculumadvice

We keep hearing that data scientist is the… Feb 27, 2018 Rachel Thomas

### Practical Deep Learning for Coders 2018courses

Last year we announced that we were developing a new deep learning course based on Pytorch (and a new library we have… Jan 26, 2018 Jeremy Howard

### International Fellowship applications for Part 2 now opencourses

From the time of our very first deep learning course at the USF Data Institute (which was recorded and formed the basis of our MOOC), we have allowed selected students that could not participate in person… Jan 17, 2018 Rachel Thomas

### New Opportunities For New Deep Learning Practitionerscourses

ai-in-society

Dawit Hailefought against the odds when he decided to study computer science in Eritrea, East Africa, despite having no internet connectivity. His perseverance paid off, first landing a job with the Eritrean government department of education, and later as an… Jan 9, 2018 Rachel Thomas

### Five Trends to Avoid When Founding a Startupadvice

This post was inspired by a round-table discussion I led on the topics of founding start-ups and personal branding at the Women in Machine Learning Workshop, co-located with deep… Jan 8, 2018 Rachel Thomas

### Deep Learning Diversity Fellowship Applications Now Opencourses

Cutting Edge Deep Learning for Coders (Part 2) will be taught this spring at the USF Data Institute in downtown San Francisco, on… Jan 2, 2018 Rachel Thomas

### Making Peace with Personal Brandingadvice

As a child, I was nerdy and shy. At my elementary and middle schools, we had to present our science… Dec 18, 2017 Rachel Thomas

### What you need to do deep learningadvice

technical

This post has been translated into Chinese here. Nov 16, 2017 Rachel Thomas

### How (and why) to create a good validation settechnical

An all-too-common scenario: a seemingly impressive machine learning model… Nov 13, 2017 Rachel Thomas

### When Data Science Destabilizes Democracy and Facilitates Genocideai-in-society

What we’re talking about is a cataclysmic change… What we’re talking about is a major foreign power with sophistication and ability to involve… Nov 2, 2017 Rachel Thomas

### Credible sources of accurate information about AIadvice

There is a lot of misleading and even false… Sep 19, 2017 Rachel Thomas

### Can Neural Nets Detect Sexual Orientation? A Data Scientist’s Perspectiveai-in-society

This post has been translated into Chinese here. Sep 13, 2017 Jeremy Howard

### Introducing Pytorch for fast.aitechnical

courses

The next fast.ai courses will be based nearly entirely on a new framework we have developed, built on Pytorch. Pytorch is a different kind of deep learning library (dynamic, rather than static), which has been adopted by… Sep 8, 2017 Jeremy Howard

### International Fellowship applications for Part 1 now opencourses

This post is from 2017. To apply for the 2018 version, please read this post Sep 8, 2017 Jeremy Howard

### Notes on state of the art techniques for language modelingtechnical

Edit one day later…… Aug 25, 2017 Jeremy Howard

### Advice to Medical Experts Interested in AIadvice

This week’s Ask-A-Data-Scientist column is from a medical doctor. Email your data science advice questions to mailto:[email protected]. Previous posts include: Aug 21, 2017 Rachel Thomas

### Sponsor a Deep Learning Diversity Scholarshipcourses

Last year’s diversity fellowships (funded by University of San Francisco and fast.ai), open to women, people of Color, LGBTQ people, and vets, played a role in helping us create a diverse community.… Aug 17, 2017 Rachel Thomas

### Diversity Crisis in AI, 2017 editionai-in-society

Deep learning… Aug 16, 2017 Rachel Thomas

### Announcing fast.ai diversity scholarshipscourses

At fast.ai, we want to do… Aug 15, 2017 Rachel Thomas

### Cutting Edge Deep Learning for Coders: Launching Deep Learning Part 2courses

Special note: we’re teaching a fully updated part 1, in person, for seven weeks from Oct 30, 2017, at the USF Data Institute. See the course page fo… Jul 28, 2017 Jeremy Howard

### Thoughts on OpenAI, reinforcement learning, and killer robotsai-in-society

“So how is fast.ai different from OpenAI?”… Jul 28, 2017 Rachel Thomas

### New fast.ai course: Computational Linear Algebracourses

I am thrilled to release fast.ai’s newest free course,Computational Linear Algebra, including an online textbook and a series of videos, and covering applications (using Python) such as how to identify the foreground in a surveillance video, how to categorize documents, the algorithm powering Google’s… Jul 17, 2017 Rachel Thomas

### How to Encourage Your Child’s Interest in Science and Techadvice

This week’s Ask-A-Data-Scientist column is from a parent on how to encourage their child in STEAM. Please email your data science related quandaries to mailto:[email protected]. Previous posts… May 7, 2017 Rachel Thomas

### Alternatives to a Degree to Prove Yourself in Deep Learningadvice

This post has been translated into Chinese here. Apr 6, 2017 Rachel Thomas

### To become a data scientist, focus on codingadvice

This week’s Ask-A-Data-Scientist column answers two short questions from students. Please email your data science related quandaries to mailto:[email protected]. Note that questions are edited for clarity and brevity.… Mar 23, 2017 Rachel Thomas

### Machine learning hasn’t been commoditized yet, but that doesn’t mean you need a PhDadvice

This post has been translated into Chinese here. Mar 17, 2017 Rachel Thomas

### How to change careers and become a data scientist - one quant’s experienceadvice

This post has been translated into Chinese here. Mar 1, 2017 Rachel Thomas

### Deep Learning: Not Just for Silicon Valleyai-in-society

courses

Recent American news events range from horrifying to dystopian, but reading the applications of our fast.ai international fellows brought me joy and optimism. I was blown away by how many bright, creative, resourceful folks from all over the world are applying deep learning to tackle a variety of… Feb 27, 2017 Rachel Thomas

### Deep Learning For Coders - Full notes and transcripts now availablecourses

We’ve been so excited to watch the thousands of people working their way through part 1 of Practical Deep Learning For Coders, and the… Feb 24, 2017

### Diversity and International Fellowships for Deep Learning Part 2courses

Applications are now open for Deep Learning Part 2, to be offered at the University of San Francisco Data Institute on Monday evenings, Feb 27-April 10. The course will cover integrating multiple cutting-edge deep learning techniques, as well as combining classic machine learning techniques with deep… Jan 28, 2017 Rachel Thomas

### Practical Deep Learning Part 2 - Integrating Recent Advances and Classic Machine Learningcourses

With part 2 of our in person SF course starting in 6 weeks, and… Jan 17, 2017 Jeremy Howard

### Big deep learning news: Google Tensorflow chooses Kerastechnical

Buried in a Reddit comment, Francois Chollet, author of Keras and AI researcher at Google, made an exciting announcement: Keras will be the first high-level library added to core TensorFlow at Google, which will effectively make it TensorFlow’s default… Jan 3, 2017 Rachel Thomas

### Where is AI/ML actually adding value at your company?ai-in-society

advice

An interesting thread came up over at Hacker News: Ask HN: Where is AI/ML actually adding value at your company?. And the folks at High Scalability were good enough to summarize… Dec 29, 2016 Jeremy Howard

### The Deep Learning MOOC is now available!courses

We’re very excited and proud to announce the launch of the fast.ai Deep Learning MOOC. It contains all the lessons from the in-person course we’ve been discussing here over the last few months, along with extra online material to help students understand the… Dec 24, 2016 Jeremy Howard

### (Update - problem resolved!) Azure and AWS’s ‘GPU general availability’ liestechnical

Huge thanks to Boyd Mcgeachie from AWS for reaching out to us and organizing a (nearly) frictionless AWS onboarding experience for our MOOC participants. He couldn’t have… Dec 19, 2016 Jeremy Howard

### So you are interested in deep learningadvice

This was inspired by a bright high school student that emailed me for advice about his interest in deep learning. Dec 19, 2016 Rachel Thomas

### How should you structure your Data Science and Engineering teams?advice

I sometimes receive emails… Dec 8, 2016 Rachel Thomas

### It used to be called big data and now it’s called deep learningtechnical

advice

At the Financial Times-Nikkei conference onThe Future of AI, Robots, and Usa few weeks ago, Andreessen… Nov 17, 2016 Rachel Thomas

### Additional Diversity Fellowship, New International Fellowships, and Deadline Extended to 10/17courses

We have been getting a lot of interest in our upcoming deep learning course over the last couple of days. With applications closing today… Oct 12, 2016 Rachel Thomas

### New Electricityai-in-society

advice

I’ve… Oct 11, 2016 Jeremy Howard

### The Diversity Crisis in AI, and fast.ai Diversity Fellowshipai-in-society

courses

Update: The deadline has been extended to 10/17. Read more here Oct 9, 2016 Rachel Thomas

### What We Will Cover in the First Deep Learning Certificatecourses

For those of you considering joining our deep… Oct 8, 2016 Jeremy Howard & Rachel Thomas

### Providing a Good Education in Deep Learningcourses

education

Paul Lockhart, a Columbia math PhD, former Brown professor, and K-12 math teacher, writes in the influential essay A Mathematician’s Lament of a nightmare world where children are not allowed to listen to or play music until they have spent over a decade mastering music notation and theory, spending classes… Oct 8, 2016 Rachel Thomas

### A unique path to deep learning expertisecourses

If you’re considering investing time in studying deep learning with fast.ai, then we want to… Oct 8, 2016 Jeremy Howard

### The First Certificate in Deep Learningcourses

Update: The deadline has been extended to 10/17 Oct 8, 2016 Jeremy Howard

### Launching fast.aiai-in-society

Jeremy is the past president of Kaggle, founder of Enlitic, FastMail.FM, and Optimal Decisions Group, and is on the faculty at Singularity University. See fast.ai’s About page for a brief… Oct 7, 2016 Jeremy Howard