# Talk Title Supercharge your AKS deployments with Wasm # Abstract Machine Learning inference (though what we talk about can be applied to other computationally intensive tasks) is often a computationally intensive task and could greatly benefit from the speed of WebAssembly. What if I told you, that you could improve the performance of your application deployed on Linux containers. Enter Wasm and Kubernetes. This talk starts off by introducing the audience to WebAssembly (Wasm) and how they could make use of the speed and security among others of Wasm for their deployments. Another problem we face is that the standard WebAssembly provides very limited access to the native OS and hardware, such as multi-core CPUs, GPU, or TPUS which is not ideal for the kind of systems we target. The talk also shows how one could use the WebAssembly System Interface (WASI) to get security, portability, and native speed for Machine Learning models. The talk then slides into the recent (still preview) support that Azure has launched for WASI node pools to run such a system while also using Krustlet to run Wasm on Kubernetes. To top it off this talk ends with a demo of doing so on Azure. # Biography I am a first year undergrad at the University of Toronto. At the moment I work on the Machine Learning team on the FINCH mission (SpaceX) which will be launched in 2024 on a Falcon 9. I love working with Machine Learning, especially Computer Vision and Kubernetes but you will also find me working with Android. I am an active contributor to multiple open-source projects like TensorFlow, KubeFlow, and Kubernetes. I also love building open-source projects (usually related to Kubernetes and Machine Learning) and some of which have been #1 trending on GitHub. Seeing my work at a rather young age, I was invited to speak at 2 TEDx and 1 TED-Ed events, I have represented my country in multiple international hackathons and won a few too and I’m also a Linux Foundation Dan Kohn Scholar. Most of the time you will find me working on a new ML research idea and I have a couple of published papers as well. I'm also a Gold Microsoft Learn Student Ambassador. Twitter: [rishit_dagli](https://twitter.com/rishit_dagli) GitHub: [Rishit-dagli](https://github.com/Rishit-dagli) LinkedIn: [rishit-dagli](https://www.linkedin.com/in/rishit-dagli/) # Image https://www.figma.com/file/eHq3jBQSXkfmV3IgFEmn2O/Rishit-Images?node-id=0%3A1 # Resources Some Talks: - KubeCon + CloudNativeCon North America: Upcoming - OSS Japan: Upcoming - Kubernetes on Edge Day NA: Upcoming - Knative Con: Upcoming - Kubeflow Summit: Upcoming - Kubernetes Communtiy Days: https://youtu.be/9rHME54q1N8?list=PLj6h78yzYM2MdrGMSTYEAVIRlYBfXd1K8 (WebAssembly based AI as a Service with Kubernetes) - OSS latin America: https://osslatam22.sched.com/event/15BrJ (Federated Machine Learning with Kubernetes), https://osslatam22.sched.com/event/15BrY (Deploying ML at Scale with Kubernetes and TFX) and https://osslatam22.sched.com/event/15BrM (WebAssembly based AI as a service) - TensorFlow Mumbai: https://youtu.be/17PAMH3ysjA (Fantastic Models and Where to find them) - Postman Summit: https://www.twitch.tv/videos/1121993724?t=01h36m00s (Deploying an ML Model as an API and using Postman to test it) - TensorFlowJS Show and Tell: https://youtu.be/IeTibm880ys?list=LL&t=4318 - TensorFlow Everywhere India: https://youtu.be/LTtgaJLo378?t=7624 (Making Deployments Easy with TF Serving) - Droidcon APAC: https://www.droidcon.com/2020/12/15/superpower-your-android-apps-with-ml-android-11/?video=491094795 (Superpower your Android Apps with ML) - DevFest: https://youtu.be/5lr1IbTpwpI?t=113 (Deploying Models to Production with TF Serving) - DevFest: https://youtu.be/m1czoyVqb1U?t=25298 Authoring and open-source: - https://link.springer.com/article/10.1007/s00500-021-05891-2 - https://arxiv.org/abs/2112.09569 - https://ieeexplore.ieee.org/document/9461296 - https://www.freecodecamp.org/news/how-to-start-an-open-source-project-on-github-tips-from-building-my-trending-repo/ - https://github.com/Rishit-dagli/Greenathon-Plant-AI - https://www.freecodecamp.org/news/graph-neural-networks-explained-with-examples/ - https://keras.io/examples/vision/swin_transformers/ - https://keras.io/examples/vision/nnclr/ - https://github.com/Rishit-dagli/MIRNet-TFJS - https://github.com/Rishit-dagli/Fast-Transformer - https://www.freecodecamp.org/news/how-to-build-better-machine-learning-models/ - https://towardsdatascience.com/debugging-your-neural-nets-and-checking-your-gradients-f4d7f55da167