Home Experience
Over four years, four teams, three clouds.
Each role taught me something different. Here’s what I owned, what I shipped, and what I took away from it.
Mar 2025 to now
Bloomington, Illinois
DevOps Engineer (AWS/GCP)
State Farm
I lead the design and implementation of a scalable, secure and automated multi-cloud infrastructure for end-to-end AI/ML work on AWS and GCP: training, deployment and monitoring.
What I own
- CI/CD pipelines on GitHub Actions, Jenkins and Buildkite.
- Infrastructure in Terraform and CloudFormation.
- EKS and GKE clusters, Helm charts, ArgoCD and FluxCD, and autoscaling for high-traffic applications.
- GPU and TPU-enabled clusters for deep learning workloads.
- Secrets management with Vault, AWS Secrets Manager and GCP Secret Manager.
- Observability with Prometheus, Grafana, CloudWatch and the ELK stack.
- Incident response and root cause analysis.
Results
- 40% shorter release cycles
- 95% fewer interruptions during releases
- 30% faster training and inference
- 25% less downtime
- 15% lower cost for GPU workloads
- 99.9% uptime
Also
I integrated SageMaker, Bedrock and Hugging Face into the DevOps pipelines and worked with data scientists and ML engineers to put LLM-based applications into production.
Read more: Faster path to production, Releases nobody notices, GPU platform for ML, Seeing problems first, Security in the pipeline
May 2022 to Jun 2023
Hyderabad, India
Cloud/DevOps Engineer
Amistad Technologies
This was my on-call education. I supported AWS services in production, automated the repetitive work around them, and learned what really matters when something breaks.
What I did
- L1/L2 support for EC2, S3 and IAM, plus networking incidents.
- Configured and troubleshot CI/CD pipelines in Jenkins, GitHub Actions and Azure DevOps.
- Automated backups, log archiving and IAM provisioning with Bash and Python scripts running on Lambda.
- Wrote reusable Terraform modules for S3 buckets, IAM roles and EC2 instances.
- Set up CloudWatch alarms, dashboards and log filters.
- Supported disaster recovery exercises: backup restores and DNS failover tests.
- Rotated IAM credentials, enforced least privilege and managed secrets with the DevOps and InfoSec teams.
Results
- 95% SLA adherence
- 50% less manual provisioning
- 20% fewer failed deploys
- 10% lower AWS cost
What I took away
A clear alert, a tested backup and a rollback you trust are worth more than any clever architecture at 2 a.m.
Mar 2021 to Feb 2022
Bengaluru, India
Cloud/DevOps Engineer
Taal Tech
My first cloud role. I worked across AWS, Azure and GCP and learned to treat infrastructure as code.
What I did
- Supported deployment and maintenance of multi-cloud environments, provisioning EC2 instances, virtual machines and storage.
- Configured IAM roles, security groups and key rotation to keep access secure and compliant.
- Wrote Shell and Python scripts for health checks, log collection and cleanup.
- Contributed Terraform and CloudFormation templates for consistent, version-controlled environments.
- Helped set up CI/CD in Jenkins and GitHub Actions to automate build, test and deploy.
- Monitored performance with CloudWatch and Azure Monitor and escalated issues early.
- Troubleshot Kubernetes pod restarts, configuration errors and resource limits.
- Wrote runbooks and onboarding material during Agile sprints.
What I took away
Once infrastructure lives in code, it can be reviewed, repeated and rolled back. I haven’t wanted to build it any other way since.
Jul 2020 to Feb 2021
Kochi, India
Linux Administrator
Ignitarium
Where it started. I kept Linux servers healthy for development teams and began scripting my way out of repetitive work.
What I did
- Set up and troubleshot Linux servers for development teams and monitored server health.
- Wrote shell scripts for user provisioning, file management, process monitoring and scheduled backups.
- Managed user permissions, cron jobs and log rotation.
- Handled patching, kernel updates, package installs and security hardening.
- Worked with developers to fix application performance bottlenecks.
What I took away
The fundamentals: permissions, processes, logs and patches. Everything in the cloud still runs on them.
Education
I did the B.Tech alongside my jobs, so the dates overlap. The about page explains it, along with the break between Taal Tech and Amistad.
-
Master of Science, Management Information Systems
University of Illinois Springfield, Springfield, IL · Dec 2024
-
Bachelor of Technology, Electronics and Communication Engineering
Aditya University, Surampalem, Andhra Pradesh, India · Apr 2023
-
Diploma, Electronics and Communication Engineering
AANM & VVRSR Polytechnic, Gudlavalleru, Andhra Pradesh, India · Apr 2020
Hiring for DevOps or platform work?
I’m open to DevOps, SRE and platform engineering roles. Email is the fastest way to reach me.