Asim Abbas
Assistant Director Devops Architect @Moody's Analytics
Signup · Get unlimited contacts
WORK HISTORY
Assistant Director Devops Architect @Moody's Analytics
London, GB
In the capacity of leading the DevOps function, my role encompassed various pivotal aspects. I was entrusted with designing and implementing the initial architecture for our platform, with a strong emphasis on Infrastructure as Code (IaC) to automate our infrastructure provisioning. Additionally, I played a pivotal role in the recruitment and nurturing of a capable DevOps team, contributing to the cultivation of a robust DevOps culture within our organization.The architectural scope extended to crafting systems hosted on Multi Public Cloud Services (AWS, Azure, GCP), spanning multiple regions and environments. This demanded a keen focus on ensuring high availability and redundancy across our services, and it accommodated various modalities such as microservices and serverless applications.Technologies experience:Terraform, Dockers, Kubernetes (EKS/AKS/GKS)/ codebuild/ cloudformation/ config/ SSM, Kubernetes, Hashicorp Vault/consul/connect.
ABOUT ASIM ABBAS
As a Lead DevOps Engineer specializing in AI Infrastructure & MLOps, I architect and implement robust, scalable platforms for developing, deploying, and monitoring machine learning models and Large Language Models (LLMs). My expertise lies in building GPU-accelerated Kubernetes ecosystems (EKS/GKE) optimized for distributed training and high-throughput inference, leveraging tools like Karpenter for efficient scaling and custom schedulers for complex workloads. I design and automate end-to-end MLOps pipelines integrating experiment tracking (MLflow), model registries, and secure, GitOps-driven CI/CD for seamless model promotion from research to production. By implementing scalable serving frameworks (KServe, Seldon Core) on Kubernetes, managing feature stores, and enforcing DevSecOps practices across the AI lifecycle, I empower researchers and data scientists. My focus is on creating resilient, multi-region AI platforms on AWS (leveraging SageMaker, P/G instances, S3) using Terraform, while fostering collaboration to streamline the path from experiment to reliable, monitored production AI services
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.