Vignesh Pamu

In the 7th century, negative numbers were used in calculations in India which was innovative because Europeans dismissed these numbers as absurd until the 17th century, when they gained widespread acceptance due to their practical use in accounting and bookkeeping.

📣 Announcing the Building Production AI on Google Cloud TPUs with JAX Technical Report! JAX has evolved from a numeric library to a complete, end-to-end ecosystem for production ML at scale. 🔹 JAX AI Stack (Core): JAX, Flax, Optax, Orbax 🔹 Scale: XLA, Pathways 🔹 Apps:…

Get your EPIC in just 15 days — fast, simple, and transparent! Apply/Update → Track via SMS → Delivered straight to your doorstep. With #ECINET, explore 40+ apps/websites on one platform Download from the Play Store or visit: 🔗 ecinet.eci.gov.in #ECI #ECIInitiatives

Trigonometry taught in class X using triangles (pic1). This is a Western MISunderstanding. In my calculus course, I teach word "trigonometry" involves a CONCEPTUAL error: it is about CIRCLES, NOT triangles. Sine and cosine are circular functions. (pic2) Aryabhata's term अर्ध…

Excited to see Coral NPU, a open Edge AI platform co-designed by @GoogleResearch and @GoogleDeepMind. Optimized to run small transformer models and LLMs on wearables with support TensorFlow, JAX, and PyTorch via IREE and TFLM compilers. We are collaborating closely to optimize…

Meet Google Sans Code — the new font meticulously crafted for coders, made by @Google! It blends geometric precision with a touch of calligraphic flair in a fixed-width design that's super easy to read, even at tiny code editor sizes. Say goodbye to squinting and hello to clear,…

🎟️ The wait is over! - The #OSFC2025 ticket sale has officially kicked off! Grab your Early Bird pass and get ready for an amazing event. tickets.osfc.io

introducing chipmunk—a training-free algorithm making ai video generation 3.7x & image gen 1.6x faster! ⚡️ our kernels for column-sparse attention are 9.3x faster than FlashAttention-3 and column-sparse GEMM is 2.5x faster vs. cuBLAS a thread on the GPU kernel optimizations 🧵

In this post I dive into MLIR, a vibrant and thriving compiler technology that is part of the LLVM Foundation. I explore its origin story, impact and also why there is so much confusion around MLIR as a compiler-y thing and MLIR as an AI thing. 👇

This is New York Central Park! Are they fools to develop a massive 900 acre eco park around concrete blocks? HCU lands or Govt lands that 400 Acres of land should be earmarked for Eco space! It’s lifeline for that area! Save HCU lands!

IICT 2025 is officially underway! We hope to make the 2025 edition bigger, better and more improved in every way. We really enjoyed the experience of organising the inaugural edition, and can’t wait to share everything that we have planned this time around. Stay tuned! Follow…

I have three Ph.D. student openings in my research group at @RutgersCS starting in Fall 2025. If you are interested in working with me on efficient algorithms and systems for LLMs, foundation models, and AI4Science, please apply at: grad.rutgers.edu/academics/prog… The deadline is…

Jules is an experimental AI-powered code agent designed to help developers get things done. 🛠️ Using Gemini 2.0, it seamlessly integrates with @GitHub and allows you to offload bug fixes and other time-consuming tasks - so you can focus on more critical issues. ↓

Our Computer Architecture course is second in series after our beginner course (Digital Design & Computer Architecture). We have updated the course with many recent developments & research. Lecture videos (Fall 2024): youtube.com/watch?v=ziMRjD… @SAFARI_ETH_CMU @ETH_en @CSatETH

Today, we’re announcing Veo 2: our state-of-the-art video generation model which produces realistic, high-quality clips from text or image prompts. 🎥 We’re also releasing an improved version of our text-to-image model, Imagen 3 - available to use in ImageFX through…

Viviane Potocnik, a PhD student at @ETH_en, dove into her research at @hipeac's Summer School on Advanced Computer Architecture and Compilation for High-performance Embedded Systems: hubs.la/Q02-ndZ00 Her research focuses on how RISC-V makes computations more efficient.

YES! Newton DID steal calculus from India! (pic1,links in ALT). Claimed it on evil doctrine of Christian discovery (= any land, knowledge belongs to first Christian to spot it.(pic2) Brits stole knowledge before they could steal wealth. (pic3)../2

EVENT ALERT!📣 🫵 Did you block your calendar? Soon we will open registrations for the upcoming @EuroHPC_JU Summit! The event will take place in Krakow, Poland from 18-20 March 2025! Don't forget to save the date & register 🔜! 🔗eurohpc-ju.europa.eu/news-events/ev…

We are happy to announce the next BLR kernel meet-up. TI has graciously offered to host us Event Details: • Date: 23rd November 2024 • Time: 9 AM - 4:30 PM • Location: Texas Instruments India Pvt. Ltd. More details: groups.google.com/g/kernel-meetu…

I am hiring for Google India (BLR/HYD) and looking for the following roles: 1. Sr. Machine Learning Engineer 2. Sr. Android Engineer 3. Application Engineer 4. Sr. Software Engineer (backend) A min 7 yoe is required, as all these are senior roles. Strong algo & design is a must.