Co-founder/CEO @cedanacorp. Previously: Engooden , @MIT_CSAIL, TL Ventures
NY
Joined February 2009
- Tweets807
- Following965
- Followers444
- Likes5.3K
Neel Master retweeted
Switching AI coding models can take 34 minutes. @cedanacorp cuts it to ~1, without losing work.
In a blog co-written with @nvidia, @neelmaster and Niranjan Ravichandra show how companies can run multiple AI coding models on a single GPU server, switching between them as needed.
The result? Less idle hardware, less wasted computing power, and more output from every GPU. cedana.com/blog/coding-model…
Neel Master retweeted
Another way the current administration is trapped in the ZERO-SUM FALLACY: The H-1B visa situation.
+ You are NOT taking a job from a citizen, rather creating jobs by increasing overal economic output. Plus these people (even vegetarians) SPEND locally.
+ The US has long been benefiting from other countries' brain drain; without it its edge wd collapse. Tesla, Von Braun, von Neumann, Fermi & Einstein did NOT each take the job of a citizen.
+ It the problem is anti-Indian racism, say it, don't play games.
Neel Master retweeted
No, a Yandex cloud malfunction did not take down disinformation/troll bots. It's a dangerous misconception. This is not how it works, not even technically. After Ukrainian drones hit Yandex data centers, a viral chart showed Russian "bot traffic" collapsing. But those bots are mostly Yandex's own search crawlers, not the info-ops social media accounts. Troll farms hide behind proxies, hijacked devices and Western clouds. They're still operating. Hired influencers are also still around. The "bots went silent" story is misinformation. Believing it is plainly dangerous. It tells people the threat is gone while influence operations keep going. Ahead of elections that complacency is what they may count on.
There are a lot of posts about companies serving billions or trillions of tokens. I work on GPU infra every day and still find it hard to build an intuitive sense scale.
So I vibe-coded a quick token calculator. Enter a token volume and pick a model. It tells you how many equivalent GPUs it takes to serve it.
You can set the GPU type, model and workload. It outputs # of GPUs with a range.
Try it out: cedana.com/resources/tokens-…
Neel Master retweeted
Excited to share our latest work with Cedana: serving a portfolio of frontier coding models on a B200 node.
TLDR: @nvidia Dynamo handles inference, NeMo Switchyard intelligently routes & escalates across models, while Cedana enables stateful model switching in ~1 min! (1/2)
More enterprises are running a portfolio of models. Easy tasks go to smaller, cheaper models, while frontier models take the hard tail.
Hosting that portfolio is a GPU utilization problem. The models don't fit on a single node together, and 9-34 min cold starts make switching between them impractical.
Excited to show our work with NVIDIA's Dynamo team. We show how to serve a portfolio of frontier coding models on a single 8x B200 node, switching models in ~1 minute with in-flight sessions intact.
cedana.com/blog/coding-model…