Kiranbir Sodhia
@PeacefulCoder
Google Engineer | Former Apple and Microsoft Engineer | Education Volunteer | Subpar Tweeter | HQ Trivia Winner
You might like
I think this is literally as close as you can get to winning a #WorldSeries without actually doing so. Unreal. That was an incredible 7-game series. Congrats to the Dodgers.
Imagine losing first authorship because you got hit by a blue shell on the last lap 💀
LLMs are injective and invertible. In our new paper, we show that different prompts always map to different embeddings, and this property can be used to recover input tokens from individual embeddings in latent space. (1/6)
The trend for $GOOGL continues. Non-Search revenue is closing in on being 50% of revenue generated.
Harvard just dropped a book on ~ML systems engineering~ and it's 100% free in PDF (it will be published in print by MIT Press). systems engineering is the hot skill companies die to see in candidates, but few books explain it in the context of ML. What it covers: → ML system…
Excited to expand our partnership with Adobe, more deeply integrating Google's latest models in their apps
At Adobe MAX keynote I love that Google DeepMind is our great partner with @elicollins speaking on the main stage. Love Nano Banana and Veo 3.1 I use every day. Two of my favorite AI models for creativity at the moment.
TPUs go brrr!
Today, we announced that we plan to expand our use of Google TPUs, securing approximately one million TPUs and more than a gigawatt of capacity in 2026.
> You do Leetcode > I do LeetGPU We are not the same bro.
Announcing the completely reimagined vLLM TPU! In collaboration with @Google, we've launched a new high-performance TPU backend unifying @PyTorch and JAX under a single lowering path for amazing performance and flexibility. 🚀 What's New? - JAX + Pytorch: Run PyTorch models on…
👀
Google officially starts selling TPUs to external customers and competes directly with Nvidia now
Excited to release new repo: nanochat! (it's among the most unhinged I've written). Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single,…
2024 evals can it count letters 🥺 can it do college stuff 🤓 are its solutions diverse 👉👈 2025 evals has it worked for 30 hours yet 🦾 has it increased gdp 📈 has it discovered novel math 🧮
Quite the contrary: We're using the language that was designed as a glue language for gluing pieces together that are written in the language(s) that were designed for peak performance. Everything working exactly as designed.
its an ironic twist of fate that the most performance intensive workloads on the planet running on eye wateringly expensive hardware are run via one of the slowest programming languages with a precarious parallelism story
1K GitHub stars in 2 days! 🔥 github.com/google/tunix
Today, we’re launching Tunix, a JAX-native LLM post-training library! With Tunix, you can do the following: - Finetuning (SFT, PEFT) - RL (GRPO, PPO, etc.) - eg.: teach your LLM to reason - Preference tuning (DPO) - Distillation, etc. For more details, refer to this blogpost:…
RL research is becoming like pretraining/modeling. This is a huge vibe shift. Most research published on RL isn't using enough compute to make many of these decisions matter as much. This is slowly shifting.
practical, modern GRPO tweaks as described in Meta's Code World Models paper
Agreed. GRPO is technically wrong.
It's delightful how easy it is to deploy working prompt injection attacks via LinkedIn
- you are - a random CS grad with 0 clue how LLMs work - get tired of people gatekeeping with big words and tiny GPUs - decide to go full monk mode - 2 years later i can explain attention mechanisms at parties and ruin them - here’s the forbidden knowledge map - top to bottom,…
United States Trends
- 1. Lakers 63.7K posts
- 2. Luka 62.4K posts
- 3. Wemby 24.2K posts
- 4. Marcus Smart 5,116 posts
- 5. #LakeShow 5,135 posts
- 6. Blazers 7,483 posts
- 7. Russ 9,441 posts
- 8. Ayton 13.6K posts
- 9. Will Richard 5,870 posts
- 10. Horford 1,754 posts
- 11. #RipCity N/A
- 12. #AmphoreusStamp 5,106 posts
- 13. Podz 2,319 posts
- 14. #AEWDynamite 19.8K posts
- 15. Champagnie 1,183 posts
- 16. Spencer Knight N/A
- 17. Kuminga 3,218 posts
- 18. Thunder 31.1K posts
- 19. Nico Harrison 1,479 posts
- 20. #dispatch 59.5K posts
You might like
-
Fola Aina
@folanski -
iconwerk
@iconwerk -
bloomerang
@bloomerangTech -
Martin Fuchs 🦊
@wahl_beobachter -
CAE
@CAEngineersLtd -
STRV
@strvcom -
Allen Wang
@creativewang -
anselmoS @anselmon.bsky.social@[email protected]
@anselmonadir -
MochaManStyle.com
@MochaManStyle -
Dror Sharon 🚀🎗️🟦
@dror_sharon -
swappie
@alex_cuciureanu
Something went wrong.
Something went wrong.