
Product Leader, Economics, Math, AI and Physics . RT🚫endorsement. Opinions here are personal. AGI LLM RecSys @Meta
Seattle, WA
Joined October 2009
- Tweets1.1K
- Following687
- Followers151
- Likes2.6K
Very proud of the work from the awesome folks! Try your use cases in Muse and leave feedback in the app or connectors which you feel didn't make sense but you need one or already working.
Muse is absolutely cooking .
1/ Taking over few bill payments & setting up Recurring Ones.
2/cleaning up my gdrive and de-duping to get more stroage.
3/ Single Snapshot Calendar for my kids ( Academic, Sports & Homework)
4/My Research Tracking
5/My Order and Deliveries
6/ Doc Appointments.
7/ Scanning Subs and getting rid of useless ones
8/ International Flight search and pricing with multi leg and route preferences ( This is crazy to next level)
@MattPRD
@alexandr_wang
@bigT_sheesh
QWEN is back! Fable, needs to permanently reset any throttling right now. This is hammering pressure from OSS MODELS.
Qwen3.8 is launching and going open-weight soon!🌐
With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5.
You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out.
Can't wait to hear what you build. Stay tuned! 🚀
Token Plan
international:qwencloud.com/pricing/token-…
China:platform.qianwenai.com/prici…
This is becoming a Wild West gun slinging fight all the time. End users are the ultimate winners.
Senthil retweeted
A super long overdue (3+ years?) post on scaling laws.
Compute is expensive. Scaling laws are a way to help us reason about the optimal compute allocation between data and model size before committing to a large run.
The post covers what scaling laws predict, how compute-optimal allocation works, why Kaplan et al. and Chinchilla disagree, and how data limits + fitting details make extrapolation tricky.
lilianweng.github.io/posts/2…
most frustrating and useless customer service experience with Uber trying to set up my teen account for the first time as soon as my teen is trying to sign up, it says account is locked. This is the first time we are trying to even sign up their account @Uber
Hanging dry at Belt 15 for 30+ mins waiting for the belt to move and baggage to come out. #alaskaairlines #seatac FAIL #seatacairport @AlaskaAir @SeaTacAirport
The Great AI Showdown: Which AI Actually Understands Network Design?
My Brief Analysis on putting AIs to battle to choose an Agent which will help for network design solution.
linkedin.com/posts/vmsenthil…
if anyone can nail usability and accessibility it's #AppleEvent #Apple . So all arm chair QB influencers and reviewers should shut up their mouth and wait. Else use their DEV account to build a custom app to see what it actually means. Random screenshots and quotes are just stupid.
Senthil retweeted
If GPU optimization and systems problems excite you, why limit your impact to a single company or lab?
Working on PyTorch allows you to ship impact to the entire AI industry!
We're hiring across experiences -- junior and senior engineers.
Read more below 👇












