New opportunity

Senior Systems Engineer - Data DevOps/MLOps

EPAM Systems · Coimbatore, India

About this role

EPAM Systems is seeking a Senior Systems Engineer with deep expertise in Data DevOps and MLOps to architect, implement, and oversee large-scale data and machine learning pipelines. The role focuses on bridging the gap between data engineering and model deployment to support enterprise-level solutions. Responsibilities: Build and manage CI/CD pipelines for data integration and ML model rollout; establish cloud-based infrastructure for data processing and model training; automate data validation, transformation, and workflow orchestration; collaborate with data scientists and product teams for seamless production integration; improve model serving, monitoring, and performance; oversee data versioning and lineage to ensure reproducibility; and enforce security standards across the pipeline lifecycle. Requirements: Bachelor's or Master's degree in Computer Science, Data Engineering, or a related field; minimum 5 years of relevant experience in Data DevOps or MLOps; proficiency in cloud platforms (Azure, AWS, or GCP); experience with Infrastructure as Code (Terraform, CloudFormation, or Ansible); strong command of containerization (Docker, Kubernetes); experience with data frameworks (Apache Spark, Databricks); programming proficiency in Python with ML libraries (Pandas, TensorFlow, PyTorch); knowledge of CI/CD tools (Jenkins, GitLab, GitHub Actions); and hands-on experience with MLOps platforms (MLflow, Kubeflow) and monitoring tools (Prometheus, Grafana). Preferred qualifications include experience with DataOps principles (Airflow, dbt), data governance (Collibra), Big Data technologies (Hadoop, Hive), and relevant cloud or data engineering certifications.

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

Skills for this role

Data DevOpsMLOpsAzureAWSGCPTerraformCloudFormationAnsibleDockerKubernetesApache SparkDatabricksPythonPandasTensorFlowPyTorchJenkinsGitLab CI/CDGitHub ActionsGitMLflowKubeflowPrometheusGrafanaAirflowdbtCollibraHadoopHive

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

Architecting Scalable CI/CD for Data and ML
The role requires building and managing end-to-end pipelines that integrate data processing with ML model deployment. You must demonstrate the ability to automate the entire lifecycle, from data ingestion and transformation to model rollout, ensuring these systems are robust and production-ready.
Infrastructure Automation and Cloud Management
You are expected to establish and maintain cloud-based infrastructure for training and processing using Infrastructure as Code (IaC) tools like Terraform or Ansible. Prioritize creating scalable, durable environments on Azure, AWS, or GCP that support high-volume data workloads.
Ensuring Reproducibility and Governance
A core duty is overseeing data versioning, lineage tracking, and experiment reproducibility. You must implement strategies that maintain data integrity and security standards, ensuring that ML models are traceable and compliant with enterprise regulatory requirements.
Cross-Functional Collaboration for Production Integration
Success depends on bridging the gap between data scientists and product teams. You must be able to translate complex ML requirements into stable production environments, improving model serving and monitoring performance through tools like Prometheus and Grafana.

Likely challenges to think through

Managing Complexity in Hybrid Pipeline Tooling
The role requires proficiency in a wide array of tools (Kubeflow, MLflow, Spark, Databricks). A significant challenge is maintaining consistency across these disparate technologies; prepare by focusing on how you standardize workflows and reduce technical debt when integrating new tools into existing enterprise pipelines.
Balancing Rapid Deployment with Model Reliability
There is an inherent tension between the need for fast ML model iteration and the requirement for stable, secure production environments. Prepare to discuss your approach to implementing automated testing and monitoring that catches performance degradation early without slowing down the development cycle.

Practice your answers

Technical practice

Designing CI/CD for ML Pipelines

What to coverExplain how you would structure a CI/CD pipeline to handle both code changes and model retraining. Discuss the differences between traditional software CI/CD and MLOps, specifically regarding data versioning and model validation steps.

Infrastructure as Code (IaC) Strategy

What to coverCompare Terraform and Ansible for managing cloud-based ML infrastructure. Describe a scenario where you would choose one over the other for provisioning scalable data processing environments.

Container Orchestration in MLOps

What to coverHow do you manage resource allocation and scaling for ML workloads using Kubernetes? Address how you handle GPU scheduling and persistent storage requirements for training jobs.

Data Lineage and Reproducibility

What to coverDescribe the importance of data lineage in a production ML environment. Explain how you would use tools like MLflow or Kubeflow to ensure that a specific model version can be traced back to its training dataset and hyperparameters.

Monitoring and Alerting for ML Models

What to coverBeyond system metrics like CPU and memory, what specific indicators do you monitor to detect model drift? Explain how you would configure Prometheus and Grafana to alert on performance degradation in a production model.

Optimizing Data Processing with Spark

What to coverDiscuss common performance bottlenecks when running large-scale data transformations in Apache Spark. How do you optimize partition strategies and memory management for heavy ML feature engineering tasks?

Workflow Orchestration with Airflow

What to coverHow do you manage dependencies between complex data transformation tasks and model training jobs? Explain the use of DAGs in Airflow to ensure reliable execution and error handling in a production pipeline.

Security in Data Pipelines

What to coverWhat strategies do you implement to enforce security across the pipeline lifecycle? Focus on secrets management, data encryption at rest and in transit, and role-based access control for ML environments.

Behavioural practice

Cross-functional Collaboration

What to coverDescribe a time you had to bridge the gap between data scientists and software engineers to deploy a model. How did you resolve conflicting priorities regarding model performance versus production stability?

Troubleshooting Production Failures

What to coverTell me about a complex issue you diagnosed within a data or ML pipeline. What was your systematic approach to identifying the root cause, and what steps did you take to prevent recurrence?

Adapting to New Technologies

What to coverGive an example of a time you had to quickly learn and implement a new MLOps tool or cloud service to meet a project requirement. How did you ensure the implementation was robust and scalable?

Balancing Technical Debt and Innovation

What to coverDescribe a situation where you had to balance the need for rapid model deployment with the need for long-term infrastructure stability. How did you communicate these trade-offs to stakeholders?

Sharpen your knowledge

Topics to prepare

MLOps Lifecycle and Reproducibility
Focus on the end-to-end management of machine learning models, specifically data versioning, lineage tracking, and experiment reproducibility. Exercise: Draft a strategy for maintaining model lineage in a production environment using tools like MLflow, ensuring that every model deployment can be traced back to its specific training dataset and code version.
CI/CD for Data and ML Pipelines
Understand the integration of automated testing and deployment for data pipelines and ML models. Exercise: Design a CI/CD workflow that incorporates automated data validation and model performance testing before triggering a production rollout, addressing how to handle rollbacks if performance metrics degrade.
Infrastructure as Code (IaC) for Data Platforms
Master the principles of provisioning cloud infrastructure consistently. Exercise: Create a Terraform script to deploy a scalable data processing environment on a cloud provider (AWS, Azure, or GCP), focusing on modularity and security best practices for data storage and compute resources.

Skills to revise

Container Orchestration with Kubernetes
Develop proficiency in deploying and managing containerized ML workloads. Exercise: Write a Kubernetes manifest to deploy a model serving container, including resource limits, liveness probes, and horizontal pod autoscaling to ensure high availability and performance.
Data Processing with Apache Spark
Gain hands-on experience with large-scale data transformation. Exercise: Build a PySpark job that performs complex data cleaning and feature engineering on a large dataset, optimizing for performance by managing partitions and caching intermediate results.
Monitoring and Alerting with Prometheus and Grafana
Implement observability for production pipelines. Exercise: Configure a Prometheus exporter to track custom metrics from an ML model (e.g., prediction latency or drift) and create a Grafana dashboard to visualize these metrics with automated alerts for threshold breaches.

Your preparation checklist

Prepare a technical project walkthrough of an end-to-end MLOps pipeline
Select a past project where you automated the lifecycle of a machine learning model. Be prepared to explain how you integrated CI/CD tools like Jenkins or GitHub Actions with containerization (Docker/Kubernetes) and how you handled model versioning using tools like MLflow or Kubeflow.
Demonstrate Infrastructure as Code (IaC) proficiency
Review your experience with Terraform, CloudFormation, or Ansible. Prepare to discuss how you have used these tools to provision and manage cloud-based infrastructure (AWS, Azure, or GCP) specifically for data processing or ML training environments.
Articulate your approach to model monitoring and reliability
Prepare to discuss how you implement observability in production. Explain your experience using Prometheus and Grafana to track model performance, data drift, or pipeline latency, and describe how you handle alerting and incident resolution for data-intensive systems.
Review data engineering and orchestration fundamentals
Refresh your knowledge on Apache Spark and Databricks for large-scale data processing. If you have experience with Airflow or dbt, prepare examples of how you have used these to orchestrate complex data transformations and ensure data lineage.
Formulate strategic questions for the interviewers
Prepare questions regarding the current maturity of their MLOps stack, the primary challenges they face in productionizing models, and how the team balances security and regulatory compliance with the need for rapid deployment cycles.

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…