Description :

Position : MLOps Engineer (4+ years)

Company Info : Prama (HQ : Chandler, AZ, USA)

Prama.ai specializes in AI-powered and Generative AI solutions across Data, Cloud, and APIs, helping businesses build intelligent platforms and scalable AI-driven products.

With 500+ projects delivered globally and 30+ engagements currently in progress, Prama.ai brings deep expertise in Data Engineering, AI/ML, and Advanced Analytics across industries such as SaaS, FinTech, Healthcare, and EdTech. We have also built 6 proprietary products, reflecting our strong product engineering DNA.

Finomics.ai is Prama.ais in-house financial intelligence and analytics platform, reinforcing our commitment to building practical, real-world solutions that deliver measurable business impact.

Prama.ai is ISO 27001 certified and SOC 2 Type II compliant, demonstrating our strong commitment to security and compliance. We are a trusted partner with AWS, Google Cloud Platform (GCP), Snowflake, and Databricks.

Headquartered in Phoenix, Prama.ai operates globally with offices in the USA (Chandler, AZ), Canada (Toronto), and India (Ahmedabad and Bangalore).

https : //www.linkedin.com/company/prama-services https : //prama.ai/

https : //www.linkedin.com/company/finomics-ai https : //finomics.ai/

Location : Ahmedabad | Bangalore | Remote

Benefits : 5 Day Working | Career Growth | Flexible working | Medical Insurance

Key Skills : Kubernetes, vLLM, ML Model deployment, CI/CD pipeline for Model tuning, Model lifecycle management in Azure

About the Role :

We are looking for a hands-on MLOps Engineer who can take ML models from notebook chaos to production reality. Youll own the end-to-end ML deployment pipeline from training orchestration to scalable inference using Kubernetes, vLLM, CI/CD automation, and Azure ML services.

If you love automating everything, optimising inference performance, and building rock-solid ML infrastructure, then this role is for you.

Key Responsibilities :

Model Deployment & Serving :

- Deploy and manage ML/LLM models in production using Kubernetes

- Implement high-performance inference pipelines using vLLM / Triton / REST & gRPC APIs

- Optimize model latency, throughput, GPU utilization and autoscaling

CI/CD for ML Pipelines :

- Build automated CI/CD pipelines for model training, tuning, validation and deployment

- Integrate GitHub Actions / Azure DevOps / Jenkins for continuous integration

- Enable automated rollback and versioning of ML models

Model Lifecycle Management :

- Manage full ML lifecycle including training, experiment tracking, model registry, deployment, monitoring and retraining

- Implement model version control, data drift detection and performance monitoring

- Automate scheduled retraining workflows

Azure Cloud & Infrastructure :

- Design and maintain ML infrastructure on Microsoft Azure (Azure ML, AKS, Blob Storage, ACR, Key Vault)

- Manage GPU clusters, networking, secrets and resource optimization

- Implement secure production-grade architecture

Monitoring & Reliability :

- Implement monitoring using Prometheus, Grafana, Azure Monitor

- Track inference metrics, system health and model performance

- Ensure high availability and fault tolerance

Required Skills :

- Core MLOps Skills

- Strong experience with Kubernetes (AKS preferred)

- Experience deploying ML/LLM models using vLLM / FastAPI / TorchServe / Triton

- Hands-on experience with CI/CD pipelines for ML workflows

- Experience with model versioning and lifecycle management tools

Cloud & DevOps :

- Strong hands-on with Azure Cloud services

- Experience with Docker, Helm, Terraform (good to have)

- Knowledge of infrastructure automation and scaling

- Programming & Frameworks

- Strong Python skills

- Experience with PyTorch / TensorFlow / HuggingFace

- REST API development for model serving