Mohan Atreya

Speed of light in real-time

Robotic pizza maker in St Louis airport

Beautiful view of the Formula-1 track in Vegas

Never deprive anyone of hope. It may be all they have!

@squidfunk Right on! Keeping a close eye on the project

@MohanAtreya Will definitely happen, we're working hard on it. Right now, we first need to fight the first wave of bugs. We tested with hundreds of projects before release, but you can't find everything. Once it stabilizes, we'll move on to the module system + work on feature parity.

... and the next Zensical release! We'll probably have one every day for the next 10 days, playing whack-a-mole with bugs as they appear, when thousands of Material users test-drive Zensical on their existing projects 😃 Compatibility madness! github.com/zensical/zensi…

Enjoyed the long form discussion with Demetrios about GPU Clouds.

Just dropped a fresh MLOps Community episode with @MohanAtreya, and it's a good one. We got into the weeds on GPU clouds, AI infra headaches, and what’s actually working in the wild right now.

The view of the S.F. Bay never ceases to amaze….

👉 docs.rafay.co/blog/2025/04/1…

Data scientists shouldn’t wait days for ML environments. With Rafay PaaS, enterprises enable self-service access to Amazon SageMaker—secure, fast, and automated. #MLOps #AWS #SageMaker #AI #SelfService #DevOps #DataScience #Rafay

Demo Video: youtube.com/watch?v=pwU7G1…

Unlock more GPU power with less cost! Check out our latest blog & video on using Nvidia’s KAI Scheduler for fractional GPUs in Kubernetes. Maximize utilization, scale smarter, and boost performance with Rafay. #GPU #Kubernetes #Nvidia #CloudNative #FractionalGPUs

Blog: docs.rafay.co/blog/2025/04/0…

For a more in-depth exploration of this subject, you can explore our blog [How to Select the Right GPU for Open Source LLMs](docs.rafay.co/blog/2025/03/1…)

Our recent blog delves into the crucial factors that demand consideration during this decision-making process. Additionally, we present a comprehensive table showcasing the Top-10 open source LLMs and the recommended Nvidia GPUs for their efficient deployment and operation.

When selecting the ideal GPU model for your preferred Open Source LLM, the ubiquitous response of "It Depends" may not always yield actionable insights.

Lastly, Part 3 sheds light on utilizing Hubble and Cilium within Rafay MKS-based Kubernetes clusters for data centers. Explore the possibilities below docs.rafay.co/blog/2025/03/0…

In Part 2, we take a closer look at the turnkey integration offered with Cilium CNI. Discover more about this integration. docs.rafay.co/blog/2025/03/0…

At Rafay, we see many customers leveraging Cilium within their Kubernetes clusters on our Rafay Platform. The primary draw for most is Cilium's robust observability features powered by Hubble.