Sumit B

About Sumit B

Senior DevOps | AI Platform & MLOps Engineer | Kubernetes (AKS, EKS, GKE)| LLM, RAG, GPU Platforms | Multi-CloudI’m a Senior DevOps and Platform Engineer with 10+ years of experience building and operating large-scale cloud and Kubernetes platforms across Azure, AWS, and GCP.Over the years, my role has evolved from traditional DevOps into owning AI platform infrastructure — enabling data science teams to deploy and run models reliably, securely, and cost-efficiently in production.What I work on todayI design and operate enterprise AI platforms on Kubernetes, focusing on the platform layer rather than model development. My work includes:Building GPU-enabled Kubernetes platforms (AKS, EKS, GKE) with autoscaling and cost controlsEnabling LLM inference and RAG-based systems using KServe, MLflow, vector databases, and FastAPIImplementing model lifecycle management (versioning, promotion, rollback) without disrupting productionDesigning CI/CD pipelines for infrastructure and AI workloads using Azure DevOps and JenkinsProviding observability and FinOps visibility (Prometheus, Grafana, Kubecost) to control GPU and cloud spendData scientists focus on model logic — I focus on making the platform reliable, scalable, secure, and affordable.Core strengthsKubernetes & Platform Engineering (AKS, EKS, GKE)AI Platform & MLOps (KServe, MLflow, RAG, FastAPI)Multi-cloud architecture (Azure, AWS, GCP)Infrastructure as Code (Terraform, Ansible)CI/CD automation (Azure DevOps, Jenkins)Observability & cost optimization (Prometheus, Grafana, FinOps)How I add valueI help organizations move from experiments to production-ready AI platforms, ensuring:predictable scalingcontrolled costs (especially GPUs)secure, repeatable deploymentsclear ownership between platform and data teamsOpen to discussing Senior DevOps, MLOps, and AI Platform engineering roles (India / Remote / Global).

Experience

1. Senior Associate

   MSCI Inc.

   Sep 2022 — Present · Pune, IN

   Cloud DevOps Engineer / Infrastructure ArchitectRole SummaryPlatform-focused DevOps engineer responsible for designing and operating cloud-native and AI platform infrastructure on Kubernetes. Own the reliability, scalability, security, and cost efficiency of enterprise platforms supporting data science and analytics teams.Key Responsibilities & ImpactDesigned and operated Kubernetes platforms (AKS, GKE) supporting production workloads, including GPU-enabled inference and data platformsBuilt and managed enterprise AI platform infrastructure, enabling LLM inference and RAG-based systems while separating platform ownership from model developmentImplemented KServe for standardized, scalable model serving and MLflow for model lifecycle management (versioning, promotion, rollback)Developed FastAPI-based orchestration services for authentication, preprocessing, and routing requests to inference servicesDesigned and maintained CI/CD pipelines using Azure DevOps and Jenkins for infrastructure and platform automationProvisioned cloud infrastructure using Terraform and Ansible, ensuring repeatable and secure deploymentsImplemented observability and monitoring using Prometheus, Grafana, and ELK to improve platform reliability and performanceEstablished FinOps practices using cost visibility tools to monitor GPU usage and optimize cloud spendCollaborated closely with data science and application teams, enabling them to focus on model logic while the platform ensured production readinessTechnologiesKubernetes (AKS, GKE) · Azure · GCP · Terraform · Azure DevOps · Jenkins · FastAPI · KServe · MLflow · Prometheus · Grafana · ELK · Python · Shell Scripting

Education

• Cdac Acts

  Diploma of Education

  2014 — 2015

• NYSS

  Bachelor's degree

  2013 — 2014

Skills

• Powerpoint
• Microsoft Excel
• Sap Bw
• Ansible
• C
• Programming
• Sql
• Git
• Microsoft Word
• Chef
• Html
• Puppet
• Jenkins
• Agile Methodologies
• Docker
• C++
• Microsoft Office
• Continuous Integration
• Core Java
• Java
• .net
• Devops
• Javascript