Monitoring the tituation... https://t.co/GbiLIgX1b2

Since its launch, there have been 60 million Copilot code reviews (and counting). 👀 As AI speeds up how fast code ships, teams are using Copilot to keep review quality high without slowing down. Here's how we've implemented your feedback and evolved Copilot code reviews over time. ⬇️ https://t.co/IDAi5TZaAQ

This is a cool, practical technique for increasing AI idea diversity by adding random priming phrases & bits of end words Similar prompts produce similar ideas, but since LLMs attend more to the start & end of inputs, this approach pushes towards novelty https://t.co/wwIPSRoFjs https://t.co/AcwGiCNzxo

Baidu releases SAMA on Hugging Face Factorizing instruction-guided video editing into semantic anchoring and motion alignment to achieve state-of-the-art performance among open-source models, remaining competitive with leading commercial systems like Kling-Omni. https://t.co/YHwYkWlHUK

Questions you’d like answered during the meeting? Submit them here: https://t.co/QCu2QCdy36

Perplexity Computer can now connect to to market research data from Pitchbook, Statista and CB Insights, everything that a VC or PE firm has access to. Enjoy! https://t.co/mEi0l2lgnC

NEW: Bill Maher—frequent foe and sometimes friend of Trump—is set to receive the Mark Twain Prize for Humor at the Kennedy Center. W/ @jon_fischer: https://t.co/5Ij3NQAcTA

chuck norris memes with the bottom caption cropped out https://t.co/JZKpfwG1kP

+2 more

chuck norris memes with the bottom caption cropped out https://t.co/JZKpfwG1kP

+2 more

Meet Codex for Students. We're offering college students in the U.S. and Canada $100 in Codex credits. Our goal is to support students to learn by building, breaking, and fixing things. https://t.co/WrOtW8E8Lk https://t.co/l2T81LgKCI

Our 26.2 release just dropped and we will be walking through it live at our next community meeting on March 23 at 10 AM PT. Plus a deep dive into variadic metaprogramming in Mojo 🔥, and a community spotlight on BlazeSeq: zero-copy, GPU-friendly FASTQ parsing built in Mojo. Check out the full details and come hang with the community: https://t.co/ThMavhW2wW

@AVMiceliBarone if it was only that language i would see your point. but it wasn’t. https://t.co/QnY146LZH1

cancel your weekend and come fix open source! you can train, build, eval a solution to deal with ai slop in open source repos. icymi, most major os repos are drowning in ai generated prs and issues. it's coming from multiple angles: - well intentioned contributors scaling too fast - students trying out ai tools and not knowing best practices - rampant bots trying to get anything merged we need a solution that allows already resource constrained maintainers to carry on doing their work, without limiting genuine contributors and/or real advancements in ai coding. let's build something that scales and enables folk to contribute more. we don't want to pull up the drawbridge. I made this dataset and pipeline from all the issues and PRs on transformers. It's updated hourly so you can get the latest versions.

You can now create and edit documents directly inline using Perplexity Computer. Simply: → Create a document → Drag over a section → Say what you want to change → It edits it right there https://t.co/V2zIMI9E5I

@maxmbeck Added ✅ https://t.co/NX2sM7aUtc Thanks again! https://t.co/dlvCgNAnLE

We've introduced dockable panels to improve navigating changes in PRs In the files changed view you can now dock → Comments → Merge status → Alerts → PR overview https://t.co/LvFt7aNxr6

New in @code Insiders: Control reasoning effort from the model picker. https://t.co/82upJE9UwP

New in @code Insiders: Control reasoning effort from the model picker. https://t.co/82upJE9UwP

2026 may be the year AI starts to truly reason about biology. AlphaFold helped close the sequence → structure gap. The next frontier is sequence → functions. Today, together with @genophoria and the team at @arcinstitute , we’re releasing BioReason-Pro — the first multimodal reasoning model for protein function prediction.

Our new open-source LiteParse comes with ready-to-use agent skills that work seamlessly with coding agents. `npx skills add run-llama/llamaparse-agent-skills --skill liteparse` ..and your agents can immediately start processing documents locally as part of their reasoning process. Here's Claude Code with liteparse enabled 💪 Documentation for LiteParse agent skills: https://t.co/VEf0zkT9wS

This is where sports, AI, and robotics meet. At the Startup Draft in Pittsburgh, @CarnegieMellon & AWS will connect founders with investors, technologists, and industry leaders shaping the future of sports innovation. Got a game-changing idea? Start scaling today with AWS Activate.

LiteParse is our free, blazing-fast document parser that you can plug into 46+ different agents - with one command 🔥 From Claude Code to OpenClaw to Cursor to Warp. Use liteparse to solve a task directly or read docs as context to write code. All you have to do is `npx skills add run-llama/llamaparse-agent-skills --skill liteparse` (we borrowed Vercel’s format) Check out our video below on using liteparse with Claude Code! Repo: https://t.co/JNER0mVcB8 Blog: https://t.co/qmpDwlkidZ

+1 more

I just saw the new codex merch, im so excited thanks for @dkundel and @ajambrosino for moving so quickly on this https://t.co/7emlCPC0TF

@singaporemacro Woah, nice! None of our Zimbabwean species had a horn quite that impressive... (Ceratogyrus sp from Chipinge, Zimbabwe, awaiting description afaik) https://t.co/vr1qd1isBG

@_inc0_ @ATinyGreenCell Terminals can display images these days :) https://t.co/5yasQ27wAB

This week, I encoded Edward Tufte's data visualization principles into an API. Then I let an AI agent try to pass it. I gave @ManusAI a CSV of women's bachelor's degree percentages across STEM fields (1970-2011) and one prompt: visualize this data. It produced a standard chart. Correct data, readable axes, nothing wrong. But a legend box instead of direct labels. No annotations calling out the rise and fall of women in Computer Science. Default colors. This is what every AI agent produces right now. So I pointed it at the Tufte Test, a quality standard I built in Truesight that checks charts against seven of Tufte's core principles. The API came back: fail on direct labeling and integrated annotations. Five other criteria passed. A quality standard gives an agent something a vague prompt never can: a precise list of exactly what to fix. Manus revised on its own. Legend box became direct endpoint labels. A subtitle surfaced the key insight. An annotation marked the Computer Science peak at 37.1% in 1983. Two prompts total from me. Everything else was autonomous. Any AI agent that can call an API could do this. What matters is the pattern: encode expert judgment once, deploy it as an API, and every AI agent in your stack builds against it. Your taste becomes infrastructure at scale instead of manual review. The Tufte Test is available as a template in Truesight if you want to try it on your own charts. Full writeup + demo video: https://t.co/Xkc9AK87xG

+1 more

between codex sessions I love having a nice can of coke thanks everyone who came to the notion x oai event as well as our build lounge in New York! https://t.co/cxgZsfCi5J

Next level. 😂 https://t.co/9jyCLRYgtT

NVIDIA just released Nemotron-Cascade 2 on Hugging Face A 30B MoE model with 3B activated parameters that achieves gold medal performance at IMO and IOI 2025. https://t.co/kNkYlhpKv8

Amazon is betting on a smarter Alexa to win back users in the UK. With Alexa+, the goal is to make interactions more natural, useful and worth returning to after years of fading engagement. But expectations are higher now, and users have plenty of alternatives. The challenge is simple: can voice assistants finally become genuinely helpful, or will they remain something people try once and forget? https://t.co/sbafcU7nJf

this guy has 29 models on huggingface at page 2 ranking. no lab behind him. no sponsorship. $2,000 from his own pocket on GPU rentals. he compressed GLM-4.7 to run on a MacBook and quantized Nemotron Super the week it dropped. all public. all free. nvidia is a trillion dollar company with hundreds of teams but they are not the ones quantizing models middle of the night and pushing them out before sunrise. if nvidia stopped tomorrow their employees stop working. people like @0xSero would not. that is the difference between a paycheck and a mission. @NVIDIAAI you talk about making AI accessible. the people actually doing it are right here. 29 models deep burning their own compute with no ask except more hardware to keep going. you do not need to build another program. just look at who is already building for you. one GPU to this man would produce more public value than a hundred internal sprints. i am not asking for charity. i am asking you to invest in someone who already proved it.

@jonkkillian @fynnso Now we know they have less than 100 million users 😆 https://t.co/n0i20btwgF