Headline

DevOps at Intel | Software Development + DevOps + Agentic AI | Immediate Joiner

About

I’m a software engineer with 2+ years of experience building cloud platforms, CI systems, and developer tools at Intel. My work focuses on DevOps and platform engineering using AWS, Kubernetes, Terraform, Python, CI/CD, and observability. At Intel, I built a provisioning service for Jenkins validation, created golden-image pipelines for consistent Windows and Linux test environments, and developed Grafana dashboards that helped teams monitor capacity and failures across more than 1,000 cloud systems. I also worked on release pipelines and developer tooling, focusing on reducing repetitive work and making issues easier to identify. I’ve deployed AI inference workloads on Kubernetes using OpenVINO and OVMS, and I’ve built with AI agents, MCP, Codex, Claude Code, and model-serving APIs. I want to continue building reliable platforms that make both traditional services and AI workloads easier to deploy, observe, and operate. I did my Master's of Engineering from BITS Pilani in Computer Science. During my time there, I co-authored a research paper published at the IEEE CLUSTER conference, where we investigated the I/O performance of BeeGFS for data-intensive HPC workloads. By provisioning a 10-node cluster from bare metal and rigorously benchmarking storage configuration parameters against diverse data access patterns, our research highlighted the need for adaptive tuning in parallel file systems bridging the gap between empirical systems research and real-world performance bottlenecks.

Location

India

AI stack

antigravity

codex

github copilot

notion

openai

Work experience

Software Development Engineer (DevOps) at Intel Corporation, Jul 2024 - Feb 2026 · 1 yr 8 mos, Remote, - Validated scalable CPU-based AI inference on a private AWS EKS/Kubernetes platform by deploying FastAPI, OpenVINO, and OVMS; provisioned the cluster with Terraform and tested ALB routing, readiness probes, metrics-server, Horizontal Pod Autoscaling (HPA), and failure recovery. - Simplified Jenkins validation by building a FastAPI provisioning service that reserved machines, deployed operating systems, and tracked request status; defined the AWS stack in Terraform and shipped it on ECS Fargate with ALB, IAM, SQS, DynamoDB, CloudWatch, and GitHub Actions. - Gave platform teams a consistent view of OneCloud and GTAX reservations by building GitHub Actions collectors that normalized provider responses and published clean telemetry to Elasticsearch. - Safeguarded Intel oneAPI releases with Jenkins CI pipelines that brought together cross-team components, acceptance tests, Coverity and BDBA security scans, and market-readiness checks. - Reduced environment drift and validation failures by standardizing Windows and Linux test environments with Packer-built golden images, AWS EC2-based WIM creation, Artifactory versioning, and repeatable OneCloud/PXE deployment. - Improved capacity planning and issue detection across 1,000+ cloud systems with Grafana SLA/SLO dashboards tracking utilization, GitHub failures, and compute lost to dirty images. - Developed and released Intel DPC++/C++ IDE extensions for Visual Studio and Eclipse, giving developers project-property controls for compiler flags; resolved MSBuild and package-loading regressions using WinDBG and trace logs, and expanded CI coverage for subsequent releases.

Graduate Technical Intern at Intel Corporation, Jan 2024 - Jun 2024 · 6 mos, Hyderabad, - Engineered a Python-based ETL pipeline to integrate OneCloud and GTAX APIs. Transformed raw telemetry into structured JSON and stored it in JFrog Artifactory and MongoDB, enabling real-time visualization on Grafana dashboards for engineering teams. - Accelerated AIOps adoption by constructing a large-scale error log dataset. Parsed and classified millions of log lines to distinguish between critical failures ("bad errors") and benign warnings ("good errors"), directly supporting model training for automated failure detection. - Maintained and optimized production-grade dashboards, resolving critical bugs to ensure accurate reporting and high availability for stakeholder reviews.

Graduate Student Researcher @ Dash Lab at Department of CSIS BITS Pilani Goa Campus, Jan 2023 - Jun 2024 · 1 yr 6 mos, Goa, India, - Cluster Deployment: Architected and deployed a 4-node Linux HPC cluster, configuring inter-node networking and storage to support distributed parallel computing workloads. - BeeGFS Implementation: Installed and configured the BeeGFS parallel file system, establishing a high-throughput distributed storage architecture for data-intensive experiments. - Performance Monitoring: Engineered a real-time health and performance monitoring solution using the ELK Stack (Elasticsearch, Logstash, Kibana), creating custom dashboards to visualize system metrics and node status. - Benchmarking & Optimization: Executed rigorous I/O benchmarking using IOR and mdtest to evaluate storage performance. Identified critical bottlenecks in default BeeGFS striping patterns that caused data distribution imbalances. - Published findings in an IEEE conference paper ("Does Varying BeeGFS Configuration Affect the I/O Performance of HPC Workloads?"), detailing how custom allocation strategies significantly improve I/O throughput compared to default configurations.

Teaching Assistant at Department of CSIS BITS Pilani Goa Campus, Jan 2023 - May 2023 · 5 mos, Goa, India, Teaching Assistant for the course of Foundations of Discrete Structures and Algorithms

Teaching Assistant at Department of CSIS BITS Pilani Goa Campus, Sep 2022 - Dec 2022 · 4 mos, North Goa, Goa, India, Teaching Assistant for the course of Foundations of Data Science (2022)

Intern at Assam Power Distribution Company Limited, Jan 2020 - Feb 2020 · 2 mos, Guwahati, Assam, India

Education

Master of Engineering - MEng, Computer Science, Birla Institute of Technology and Science, Pilani

Bachelor of Engineering, Computational Science, Assam Engineering College

Public links

github: https://github.com/callmetushar123