Paras Mamgain
Engineer
New Delhi, India
Actions
Paras is a Technical Lead Manager at Google Cloud, where he leads a team dedicated to simplifying complex cloud solutions for customers. Drawing on his strong foundation in cloud solutions and backend development, he guides his team in architecting scalable and resilient infrastructure. Paras is also a passionate educator and regular international speaker at premier events like the Linux Foundation Open Source Summit, known for his ability to make advanced technology accessible to a wide audience.
Badges
Area of Expertise
Topics
Linux Foundation AI_dev: Open Source GenAI & ML Summit Europe 2025
Imagine deploying a complex AI model for real-time inference, but facing latency that slows down your application. We've all been there. This talk isn't just about theoretical serverless benefits; it's about real-world performance gains. We'll show you how we slashed inference latency from several seconds to under 100 milliseconds, achieving a 10x improvement in throughput, by harnessing the power of serverless on Kubernetes.
We'll dive into practical examples of using Knative and Kubeless to deploy containerized machine learning models as serverless functions by demonstrating the deployment of entire infrastructure provisioning process using IaC. Learn how to automate data preprocessing pipelines, reducing data preparation time by 30% through event-driven triggers.
You'll learn how to distribute model training jobs across ephemeral pods, cutting training times by 40% on a large dataset, and how to scale inference endpoints dynamically based on real-time traffic spikes.
HashiTalks: India
Cross-cloud network automation with Waypoint and Terraform
As cloud professionals, we face daily challenges in managing complex, multi-cloud infrastructures. Ensuring high availability, scaling with ease, and maintaining seamless workflows can feel overwhelming. This session will dive into how HashiCorp’s Waypoint and Terraform can make your life easier by automating and standardizing cross-cloud deployments—letting you focus on building and optimizing your applications, not managing infrastructure headaches.
https://www.youtube.com/watch?v=nXtne5SCnkQ
The Linux Foundation open Source Summit
eBPF-Powered Observability for Sustainable Computing - Paras Mamgain, Anmol Krishan Sachdeva, Samrat Priyadarshi & Miki Katsuragi, Google
With the advancements in the fields of Applied AI/ML, computational power, FinOps practices, and cloud services, organizations are increasingly building and deploying AI/ML workloads on Kubernetes. Cloud Computing gives the impression that the underlying resources (like Virtual Machines, CPU, RAM, Storage, etc.) are infinite, and hence many individuals don’t focus on the count of machines and resources when training their AI/ML models and proceed with training fairly large models with millions and billions of parameters. The energy-intensive nature of model training and iterations, coupled with the growing scale of AI/ML deployments, is contributing to a considerable environmental impact in the form of Carbon Emissions.
This talk introduces some key open-source cloud agnostic tools and techniques to build eBPF-powered observability solutions for practicing Sustainable Computing. By analyzing and tuning performance and energy metrics from Kubernetes resources like Pods and Nodes, AI/ML Practitioners and Platform Engineers can optimize workloads, reduce energy consumption, lower costs, and minimize environmental impact while maintaining performance.
https://www.youtube.com/watch?v=eGatqRtotfs
Open Source Summit Linux Foundation
Effective Observability with Open Telemetry - Paras Mamgain & Kiran Darbha, IBM
Open Telemetry provides libraries, agents and various other components that you need to capture telemetry from your services so that you can better observe, manage and debug them.
It captures metrics, distributed traces, resource metadata and the logs from your backend and client applications and sends it to the backend of Prometheus, Jaeger etc.
https://www.youtube.com/watch?v=ETXTAbuzaVw&t=20s
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top