@ben_burtenshaw @TheZachMueller What are some good agent skills for learning?
English
Mikey Zhong
1.9K posts
Mikey Zhong
@mzhong21
Interested in computer vision, transportation without cars, and accessible education.
New York, NY Katılım Mart 2014
426 Takip Edilen121 Takipçiler
Still experimenting tbh, but we know that youtube is getting (even) more impactful, and agent skills for learning are becoming a thing. The next course will try to fuse this, so basically teach ML through context engineering, and guide the course via YouTube live streams. Wish me luck!
English
Hey Ben, where are all the courses?
In 2026, I’ve cut down a lot on new courses for Hugging Face. Mainly because it felt like learning had undergone a major change, and people just weren’t upskilling in the same way.
A lot of students are learning with their agents plus some YoutTube. And I wanted to figure out a way to meet this audience. (Check out the Hugging Face YouTube channel for some great content there)
Coming soon! A new course on Hugging Face that is fully agent integrated, based on long form videos, and use real projects.
English
@GergelyOrosz I switched to graphite (stacked diffs) in 2023 and switched back to regular PRs last month since stacked diffs don't play nicely with git worktrees
English
Better late than never, but stacked diffs feels like it was relevant pre-2024/2025 for eng teams big on using AI: before AI agents became big, and started to generate large code changes.
I'd go as far to say that stacked diffs make most sense when writing a lot of code by hand
English
is there any reason you wouldn't default to having just one code tool for mcps now?
Cloudflare@Cloudflare
We’re introducing Dynamic Workers, which allow you to execute AI-generated code in secure, lightweight isolates. This approach is 100 times faster than traditional containers. cfl.re/4c2NvPl
English
btw their supabase storage bucket is publicly accessible via any signed url token 😭
exposes:
> employee background checks
> equity vesting schedules and grant amounts
> performance reviews
> session tokens for stripe, notion, etc
> screenshots below 🧵
i also got access to their notion 😛
erin griffith@eringriffith
A detailed and brutal look at the tactics of buzzy AI compliance startup Delve "Delve built a machine designed to make clients complicit without their knowledge, to manufacture plausible deniability while producing exactly the opposite." substack.com/home/post/p-19…
English
@saranormous @karpathy @NoPriorsPod His thoughts on what's needed to actualize his AI personalized tutor vision / what's the latest with Eureka Labs
English
excited to have sensei @karpathy back on @nopriorspod tomorrow. what wisdom shall we glean? (yes, yes, wth is going on with coding agents)
English
@iamRezaSayar @physical_int Yes have this question as well! Love the join text for longish-term memory, video encoder for short-term split.
English
@physical_int Awesome work! 🔥🔥👏May I ask what made you decide to train your own video encoder, as opposed to using the available ones? 👀
English
We’ve developed a memory system for our models that provides both short-term visual memory and long-term semantic memory.
Our approach allows us to train robots to perform long and complex tasks, like cleaning up a kitchen or preparing a grilled cheese sandwich from scratch 👇
English
Mercury 2 is live 🚀🚀
The world’s first reasoning diffusion LLM, delivering 5x faster performance than leading speed-optimized LLMs.
Watching the team turn years of research into a real product never gets old, and I’m incredibly proud of what we’ve built.
We’re just getting started on what diffusion can do for language.
English
@StefanoErmon wow, this is incredible and I've requested access. I've been doing some work with using LLMs to generate web animations and the tokens/sec has been the bottleneck.
English
English
visual-json for @code now in beta
Feedback welcome
github.com/vercel-labs/vi…
Chris Tate@ctatedev
Introducing visual-json > JSON editing with human-first ergonomics – Minimalist – Embeddable – Schema-aware – Extensible – Drag and drop – Keyboard navigation – Tree view to drill into deeply nested data
English
Can’t believe they didn’t go with:
Sonnet 4.5 (new)
Claude@claudeai
This is Claude Sonnet 4.6: our most capable Sonnet model yet. It’s a full upgrade across coding, computer use, long-context reasoning, agent planning, knowledge work, and design. It also features a 1M token context window in beta.
English
Mikey Zhong retweetledi
This is really cool! Including intermediate variable states as context to represent a code world model. One of the more inspiring papers I've read in a while.
Gabriel Synnaeve@syhw
(🧵) Today, we release Meta Code World Model (CWM), a 32-billion-parameter dense LLM that enables novel research on improving code generation through agentic reasoning and planning with world models. ai.meta.com/research/publi…
English
minor correction - candidates were given nanodevin, a checkpoint of early 2024 devin, and they cna improve it in whatever dimension they are passionate about. starting things from scratch would result in repetitive low value submissions and not convey some assumptions that are better communicated in code than english
English
Think I’m going to steal @cognition’s interview process.
To vet engineers for Devin, they have candidates build Devin (read: an AI engineer) from scratch. They have 8 hours to get as much done as possible.
Perfect test for system design, ability to create leverage with AI, and knowing which tools to use when.
English
not bringing advil on an international trip is the height of hubris
English
@fpawn though even if white doesn't play the winning move, your position is still so depressing. Impressed you salvaged it.
English
My weekend would have gone differently had my opponent found a cute yet simple trick. White to move. Sadly, I was black and did not see it. Fortunately, he missed it too.
English
wow this is incredible
Jianyuan@jianyuan_wang
Introducing VGGT (CVPR'25), a feedforward Transformer that directly infers all key 3D attributes from one, a few, or hundreds of images, in seconds! No expensive optimization needed, yet delivers SOTA results for: ✅ Camera Pose Estimation ✅ Multi-view Depth Estimation ✅ Dense Point Cloud Reconstruction ✅ Point Tracking Project Page: vgg-t.github.io Code & Weights: github.com/facebookresear…
English