Itamar

Generative AI @Snap

[Itamar @itamarbe]

New post! Unpredictability is Human https://t.co/raawYX9ZbL

[Itamar @itamarbe]

Thinking a bit what we can never fully delegate to AI https://t.co/OFV8SUuIxG

[Itamar @itamarbe]

New post includes Steve.md. Not for your agent. For you.

https://t.co/YHoFbtB04N

[Itamar @itamarbe]

The Suit That Changed the Race https://t.co/UJBtYWSsIR

[Itamar @itamarbe]

AI changed what execution looks like, what impact means, and what makes teams actually win. Started writing about it.

https://t.co/qgfmts6VyI

[Or Patashnik @OPatashnik]

Kontinous Kontext code is now available!

[Simon Willison @simonw]

Claude Skills are awesome, maybe a bigger deal than MCP
https://t.co/1wIYcTFrzI

[Yael Vinker @YVinker]

Join us tomorrow! ✏️🗓️
2:00pm - 5:15pm PDT 
East Building, Room 11

Looking forward to seeing everyone and learn more about drawing and sketching!🤗🎨

[Andrej Karpathy @karpathy]

+1 for "context engineering" over "prompt engineering".

People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step. Science because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting... Too little or of the wrong form and the LLM doesn't have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial. And art because of the guiding intuition around LLM psychology of people spirits.

On top of context engineering itself, an LLM app has to:
- break up problems just right into control flows
- pack the context windows just right
- dispatch calls to LLMs of the right kind and capability
- handle generation-verification UIUX flows
- a lot more - guardrails, security, evals, parallelism, prefetching, ...

So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term "ChatGPT wrapper" is tired and really, really wrong.

[Omer Dahary @OmerDahary]

Excited to share that our new work, Be Decisive, has been accepted to SIGGRAPH!
We improve multi-subject generation by extracting a layout directly from noise, resulting in more diverse and accurate compositions.
Website: https://t.co/CMhkBHygpf
Paper: https://t.co/xgP024jkn5

[Elad Richardson @EladRichardson]

Ever stared at a set of shapes and thought: 'These could be something… but what?'

Designed for visual ideation, PiT takes a set of concepts and interprets them as parts within a target domain, assembling them together while also sampling missing parts.

https://t.co/uOHWTQvhIo

[Min Choi @minchoi]

This is insane.

Meet Blender MCP, Claude AI can now talk directly to Blender.

Instantly turn any prompt or 2D image into stunning 3D scenes.

Creativity unlocked.👇

[Matan Cohen-Grumi @MatanCohenGrumi]

Pikadditions and Pikaswaps are now in 1080p Pro mode and it's sooooo good.

[Daniel Cohen-Or @DanielCohenOr1]

Vectorization into a neat SVG!🎨✨
Instead of generating a messy SVG (left), we produce a structured, compact representation (right) - enhancing usability for editing and modification. Accepted to #CVPR2025 !

[Ryo Lu @ryolu_]

this guy rebuilt what i worked on for 5 years in 8 days vibing in @cursor_ai 🥲

[GitHub @github]

"Incoming transmission!" 📡

Classic Command and Conquer games are now open source—ready for modding, preservation, and innovation!

Read more: https://t.co/f7lMwE8u5T
Dive into the code: https://t.co/O6smxWUw1i

[Yael Vinker @YVinker]

SketchAgent has been accepted to #CVPR2025 !
This is an early step toward new tools for visual thinking and richer interaction with LLMs 🎨

🔗 https://t.co/lnlLFHG8Qp

[Omri Avrahami @OmriAvr]

I'm happy to share that Stable Flow has been accepted to #CVPR2025 🥳
Source code is available at https://t.co/03ZVzf5MRn

[Alessio @alessiograncini]

Watching Live NBA on @Spectacles

[Lens Studio @LensStudioDev]

Developer Mike Berry is fairly new to creating Lenses and using Lens Studio. Over the past year, he’s used Easy Lens Beta to help him bring his ideas to life. He’s also got some advice for new creators like him. Read on for our full conversation: https://t.co/i1BZSnd3pp