MLOps Engineer · DevOps · India

Ajeet Singh.

MLOps Platform Engineer ← DevOps & CI/CD ← Oracle DBA

Turning experimental models into production-ready systems — from pipeline orchestration to scalable cloud-native ML infrastructure.

View Projects Get In Touch
Open to MLOps consultation

DevOps · MLOps Focus

  • Kubernetes & Docker — container orchestration & packaging
  • GitHub Actions & ArgoCD — CI/CD & GitOps pipelines
  • Terraform & IaC — infrastructure automation on AWS
  • AWS SageMaker — ML training & deployment
  • MLflow & Kubeflow — experiment tracking & pipelines

Core Stack

KubernetesDocker TerraformGit GitHub ActionsArgoCD AWS EKSMLflow SageMakerPrometheus AirflowOracle
Featured Work

Projects

01 / 04

AWS ML Platform-as-Code

Automated provisioning of a complete SageMaker environment using Terraform. Includes VPC networking, IAM least-privilege roles, and EKS cluster setup for distributed training.

TerraformSageMaker EKSVPC
Source Code
02 / 04

Feature Store

Low-latency Feature Store using AWS Glue and Redis. Migrates legacy Oracle relational data into versioned format for real-time ML inference with DVC for data versioning.

AWS GlueRedis DVCOracle
Source Code
03 / 04

Automated Model Retraining Loop

Full CI/CD/CT pipeline using GitHub Actions to trigger model retraining in MLflow when new data arrives in S3, with zero-downtime deployment via ArgoCD.

GitHub ActionsMLflow ArgoCDS3
Source Code
04 / 04

Model Health Dashboard

Prometheus & Grafana stack monitoring model drift. Tracks prediction latency and accuracy decay — applying DBA-style health checks to production ML systems.

PrometheusGrafana Model DriftSLOs
Source Code
About

Who I Am

I'm an MLOps & ML Platform Engineer with roots in DevOps and Oracle PeopleSoft DBA. That infrastructure background gives me a systems-level mindset — reliability, observability, and scale come before model accuracy.

My philosophy: the hardest part of ML isn't building the model — it's keeping it healthy in production. I apply DBA-level data integrity rigor to Feature Stores and DVC pipelines.

  • Infrastructure as Code: Eliminating "works on my laptop" syndrome in ML environments.
  • Data Integrity: DBA-level rigor applied to Feature Stores and data versioning.
  • Observability: Monitoring model drift and performance decay, not just system health.
5+Years Cloud Infra
3Cloud Providers
∞Pipeline Iterations
01Goal: Prod-Ready ML
Technical Expertise

Skills

MLOps & DevOps Engineering

DevOps Engineering92%

Kubernetes Docker Git GitHub Actions ArgoCD Jenkins

ML Infrastructure85%

SageMaker · MLflow · Kubeflow · Feature Stores · DVC

Data Engineering88%

PySpark · Airflow · AWS Glue · Redshift · SQL

Cloud & DevOps

Cloud & IaC90%

AWS · Terraform · EKS · CloudWatch · S3 · IAM

DevOps & CI/CD85%

Docker · Kubernetes · GitHub Actions · ArgoCD · Jenkins

Database Administration88%

Oracle · PeopleSoft · Exadata · PostgreSQL · Redis

Contact

Let's Connect

Open to MLOps and data architecture consultation, collaboration on open-source ML tooling, and conversations about industrialising AI in production.