Sujal Sahu

Cloud DevOps Engineer | Full Stack Developer | Winner @SIH\'22

Role
Devops Engineer at Appperfect Corp
Location
Udaipur, RJ, IN
LinkedIn followers
500 followers
Information TechnologyView LinkedIn profile

About Sujal Sahu

Certified Kubernetes Administrator and Cloud DevOps Engineer with ~3 years of experience…

Experience

  1. Devops Engineer

    Appperfect Corp

    Aug 2023 — Present · Udaipur, IN

    AWS HPC Parallel Cluster with Slurm - Collaborated with team to automate cluster creation process with all necessary software stack deployment via pipeline utilizing Pulumi (IAC)- Leveraged runbooks to automate cluster validation, run NVIDIA NCCL-tests and train the GPT3-7B model- Setup Monitoring using Prometheus and Grafana to capture network (EFA), node and DCGM (GPU related) metrics- Successfully trained LLM model with 355B parameters on cluster. Azure - Automated CycleCloud (HPC Slurm Cluster) deployment using terraform as IAC- Created scripts to validate the cluster and fix known issues using Ansible- Evaluated and compared various Azure services and storage solutions to utilize them to their full potential.On-Prem K8s HPC Clusters - Developed Go application etcd-manager to automate compaction & defragmentation, keep etcd size within limits and take snapshots at regular intervals- Contributed to the ArgoCD automation to manage access on controller nodes- Created ansible playbooks to automate the resolution of known issues on cluster.Nebius sOperator - Validating the performance of the sOperator cluster on Nebius with H100 GPUs and setting up observability.(sOperator - Slurm cluster running on K8s)- Evaluating the Slurm cluster build process on the top of K8s across all CSPs using the Nebius soperator.

Education

  • College of Technology and Engineering

    Bachelor's degree, Computer Science

    2020 — 2024

Find verified contacts for anyone on LinkedIn

Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.

Free plan included · No credit card required

This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.

Sujal Sahu — Devops Engineer at Appperfect Corp in Udaipur, RJ, IN | Unifers