DevOps Engineer | MLOps Engineer | AWS • Kubernetes • Terraform • CI/CD • GitOps • Python
I build automated cloud infrastructure, reliable delivery pipelines, and reproducible environments for software and machine-learning workloads. My background combines AWS/Kubernetes platform engineering, infrastructure as code, CI/CD and GitOps, observability, and MLOps deployment workflows.
My recent work at Handshake AI and AfterQuery extends that experience into evaluating coding agents on infrastructure recovery, configuration repair, Linux troubleshooting, and ML workflows.
| Project | What you can inspect |
|---|---|
| AWS EKS Platform Blueprint | A reference platform with Terraform VPC/EKS configuration, IAM boundaries, Kubernetes manifests, GitOps configuration, observability values, and operating runbooks. |
| DevOps Policy Audit Toolkit | A Python CLI for infrastructure and CI policy checks, with configurable severity gates, finding baselines, SARIF/HTML/JSON reporting, and automated tests. |
| Rental Price MLOps Pipeline | A rental-prediction case study with data validation, model quality gates, artifact and registry metadata, FastAPI serving, Docker, tests, and Kubernetes/Helm manifests. |
| ML Platform Infrastructure on Kubernetes | A model-serving infrastructure demo with a Python scoring API, Docker, Helm, Argo CD manifests, Prometheus metrics, and container smoke checks. |
These are portfolio implementations, demos, and blueprints. Each repository documents its scope, setup, and validation steps.
| Area | Tools and practices |
|---|---|
| Cloud and infrastructure as code | AWS, EKS, EC2, VPC, IAM, S3, RDS, Terraform, Terragrunt, CloudFormation, Ansible; working exposure to Azure and GCP |
| Containers, CI/CD, and GitOps | Docker, Kubernetes, Helm, Kustomize, GitHub Actions, GitLab CI/CD, Jenkins, Argo CD, Flux CD |
| MLOps and model deployment | MLflow, Kubeflow Pipelines, FastAPI, model artifacts, experiment tracking, quality reports, drift checks |
| Automation and testing | Python, Bash, PowerShell, SQL, pytest, Git, Linux, automated regression testing |
| Observability, reliability, and security | Prometheus, Grafana, CloudWatch, ELK/EFK, Datadog, IAM, OIDC, RBAC, KMS, Secrets Manager, incident response, RCA |
| AI evaluation | Agent benchmarks, reference solutions, hidden verifiers, trajectory analysis, mutation testing |
| Data infrastructure | Kafka, Apache Spark, dbt, Amazon Redshift, Snowflake |
At Handshake AI (Jun-Sep 2026), my work included:
- Developing coding and terminal-agent benchmarks covering systems infrastructure, software engineering, data processing, and machine learning.
- Building reproducible Docker environments, Python/Bash reference solutions, and independent hidden verifiers, with Oracle, NOP, regression, and mutation testing.
- Investigating agent execution and infrastructure failures, and using pass@2/pass@5 results alongside execution trajectories to identify shortcuts and reliability issues.
- Consolidating 95 repositories across 15 categories and 52 subcategory groups into a private GitHub library, preserving source-commit provenance and automated integrity checks.
- Verifying 5,440 files, preserving executable permissions, and reviewing secret-scan findings.
Client benchmark implementations remain private.
At AfterQuery (Jun-Jul 2026), I built repository and terminal-based evaluations for configuration repair, Linux troubleshooting, service recovery, bug fixing, and refactoring, with fail-to-pass and pass-to-pass regression checks.
- Terraform GitOps Delivery Platform
- SRE Observability and Incident Response Lab
- Secure Kubernetes Platform Hardening
- Cloud Cost Optimization Toolkit
Based in Delhi NCR, India. Interested primarily in DevOps Engineer, MLOps Engineer, and Cloud Platform Engineer opportunities, with a focus on AWS, Kubernetes, infrastructure automation, CI/CD, GitOps, and reliable ML infrastructure.
I also bring coding-agent evaluation experience relevant to AI infrastructure and evaluation engineering roles. My research interests include multimodal and multilingual model evaluation and agentic systems.