New opportunity

Senior Systems Engineer - Data DevOps/MLOps

EPAM Systems · Bangalore, India

About this role

EPAM Systems is seeking a Senior Systems Engineer specializing in Data DevOps and MLOps to join their team. The role focuses on building enterprise-level solutions by integrating machine learning models into live production environments and managing scalable data pipelines. Responsibilities include designing CI/CD pipelines for data and ML model deployment, maintaining cloud-based infrastructure for model training, and automating data validation and orchestration workflows. You will collaborate with cross-functional teams to ensure model reproducibility, performance, and reliability while maintaining data lineage and security compliance. Requirements include a Bachelor's or Master's degree in Computer Science or a related field and at least 5 years of experience in Data DevOps or MLOps. Candidates must have hands-on experience with cloud platforms (Azure, AWS, or GCP), Infrastructure as Code (Terraform, Ansible), containerization (Docker, Kubernetes), and data processing frameworks (Apache Spark, Databricks). Proficiency in Python and familiarity with ML libraries (TensorFlow, PyTorch, Pandas) and MLOps platforms (MLflow, Kubeflow) are essential. Preferred qualifications include exposure to DataOps tools like Airflow or dbt, knowledge of data governance frameworks like Collibra, and experience with Big Data technologies such as Hadoop or Hive. The position is offered in a hybrid work mode across multiple Indian locations.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

Data DevOpsMLOpsMLflowKubeflowCI/CDInfrastructureDockerKubernetesApache SparkDatabricksPythonPandasTensorFlowPyTorchJenkinsGitGitHub ActionsTerraformAnsibleData AnalysisCloud ComputingAzureAWSGCPPrometheusGrafana

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

End-to-End MLOps Pipeline Orchestration
The core requirement is to design and maintain CI/CD pipelines that bridge the gap between data science experimentation and production deployment. You must demonstrate the ability to automate model training, validation, and serving while ensuring reproducibility and lineage tracking using tools like MLflow or Kubeflow.
Scalable Cloud Infrastructure Management
You are expected to deploy and manage cloud-based infrastructure (AWS, Azure, or GCP) using Infrastructure as Code (IaC) principles. This involves utilizing Terraform or Ansible to ensure that data processing environments and model training clusters are resilient, secure, and performant.
Cross-Functional Collaboration and Reliability
Success in this role requires working closely with data scientists and software engineers to translate experimental models into reliable production services. You must prioritize system monitoring, logging, and alerting to maintain high availability and performance standards.

Likely challenges to think through

Balancing Model Agility with Production Stability
A common challenge in MLOps is the friction between the rapid iteration cycles of data scientists and the strict stability requirements of production environments. Prepare to discuss how you implement automated testing and validation gates to prevent model drift or pipeline failures without stifling development speed.
Managing Complex Data Lineage and Governance
Maintaining data integrity and compliance across distributed pipelines is difficult. You may face challenges in ensuring that data versioning and lineage are strictly tracked. Be ready to explain how you integrate governance frameworks and automated validation to ensure security and auditability.

Practice your answers

Technical practice

Designing CI/CD for ML Models

What to coverExplain how you would structure a CI/CD pipeline specifically for ML model deployment compared to traditional software. Focus on how you handle model versioning, data lineage, and automated testing for both code and model artifacts.

Infrastructure as Code (IaC) for MLOps

What to coverDescribe your experience using Terraform or Ansible to provision and manage cloud infrastructure for ML workloads. Discuss how you ensure infrastructure consistency and scalability when deploying environments for model training and serving.

Containerization and Orchestration in Production

What to coverHow do you leverage Docker and Kubernetes to ensure the reliability and scalability of ML model serving? Address challenges related to resource management, auto-scaling, and maintaining environment parity between development and production.

Data Processing and Orchestration

What to coverDiscuss your approach to automating data validation and transformation pipelines using frameworks like Apache Spark or Databricks. Explain how you integrate these with orchestration tools to ensure data quality and timely delivery for model training.

MLOps Platform Implementation

What to coverCompare the use of MLflow and Kubeflow in an enterprise environment. Explain how you would use these tools to manage the end-to-end lifecycle of a model, including experiment tracking, registry, and deployment.

Monitoring and Reliability

What to coverHow do you utilize tools like Prometheus and Grafana to monitor ML model performance in production? Describe how you set up alerting for model drift or infrastructure failures to ensure system reliability.

Data Governance and Security

What to coverExplain the importance of data lineage and security compliance in an MLOps pipeline. Discuss how you would implement protocols to protect data integrity and ensure that models remain reproducible and compliant with governance standards.

Troubleshooting Pipeline Failures

What to coverDescribe a scenario where you had to troubleshoot a failure in a complex data or ML pipeline. Detail your diagnostic process, the tools you used to identify the root cause, and how you implemented a permanent fix.

Behavioural practice

Cross-functional Collaboration

What to coverDescribe a time you collaborated with data scientists and software engineers to move an ML model into production. How did you align the different technical requirements and priorities of these teams to ensure a successful deployment?

Managing Technical Debt in Infrastructure

What to coverProvide an example of when you identified an opportunity to improve deployment processes or infrastructure resilience. Explain the situation, the actions you took to implement the improvement, and the resulting impact on the team's efficiency.

Handling Ambiguity and Independent Problem Solving

What to coverTell me about a time you had to solve a complex technical issue independently with limited guidance. How did you approach the problem, and what steps did you take to ensure your solution was documented and communicated effectively?

Adapting to Changing Project Requirements

What to coverDescribe a situation where project requirements shifted during the development of a data pipeline. How did you adjust your technical approach while maintaining the integrity and performance of the system?

Sharpen your knowledge

Topics to prepare

MLOps Lifecycle and Model Reproducibility
Focus on the end-to-end process of moving models from experimentation to production. Review strategies for maintaining data lineage, versioning, and ensuring model reproducibility. Exercise: Map out a CI/CD pipeline architecture for an ML model that includes automated testing, validation, and deployment stages, specifically addressing how to handle model drift and retraining triggers.
Infrastructure as Code (IaC) and Cloud Orchestration
Understand the principles of managing cloud-based infrastructure through code to ensure scalability and resilience. Review how to provision and manage resources on AWS, Azure, or GCP using Terraform or Ansible. Exercise: Create a Terraform script to provision a scalable Kubernetes cluster and a managed data processing service like Databricks, focusing on security and cost-optimization best practices.

Skills to revise

Containerization and Orchestration with Kubernetes
Master the deployment and management of containerized applications. Exercise: Containerize a Python-based ML inference service using Docker and define a Kubernetes deployment manifest that includes horizontal pod autoscaling and health checks to ensure high availability.
MLOps Platform Implementation (MLflow/Kubeflow)
Gain hands-on experience with tools that manage the ML lifecycle. Exercise: Use MLflow to track a model training experiment, log parameters and metrics, and register a model version. Then, demonstrate how to deploy that registered model into a staging environment using a CI/CD pipeline.
Data Pipeline Automation with Apache Spark
Develop skills in processing large-scale datasets efficiently. Exercise: Build a PySpark job that performs data validation and transformation on a raw dataset, ensuring the output is partitioned correctly for downstream consumption in a production environment.

Your preparation checklist

Prepare a technical walkthrough of an MLOps pipeline
Develop a concise presentation or document detailing a past project where you automated the deployment of an ML model. Highlight your use of CI/CD tools, containerization (Docker/Kubernetes), and how you ensured model reproducibility and data lineage.
Demonstrate Infrastructure as Code (IaC) proficiency
Prepare to discuss specific scenarios where you used Terraform or Ansible to provision cloud infrastructure (AWS, Azure, or GCP). Be ready to explain how you managed state, handled configuration drift, and ensured security compliance in your code.
Review data orchestration and processing frameworks
Refresh your knowledge on Apache Spark and Databricks. Prepare to explain how you have optimized data pipelines for performance and how you integrated these frameworks with MLOps platforms like MLflow or Kubeflow.
Formulate questions regarding EPAM's MLOps maturity
Prepare thoughtful questions for the interviewers about their current stack, such as how they handle model monitoring in production using tools like Prometheus or Grafana, and how they approach data governance and security at scale.
Refine your Python and ML library expertise
Be prepared to discuss your experience using Python, Pandas, TensorFlow, or PyTorch in a production environment. Focus on how you write maintainable, testable code for data transformation and model serving.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…