MLOPS Engineer

Experience: 4-15Years

Location: Pan India(Metro Cities)

Job Title: ML Ops Engineer

Responsibilities:

1. Model Deployment & Integration:

• Design, develop, and manage automated pipelines for deploying machine learning models into production.

• Ensure smooth integration between model development, data, and application teams.

• Implement model versioning and rollback strategies to facilitate easy model updates and troubleshooting.

2. Infrastructure Automation:

• Build and maintain scalable infrastructure using tools like Kubernetes, Docker, and cloud platforms (AWS, Azure, GCP).

• Automate the deployment process and manage model serving environments.

• Design and optimize cloud-native solutions to ensure scalability and performance under heavy workloads.

3. Monitoring and Maintenance:

• Continuously monitor the performance and health of deployed models in production environments.

• Implement real-time logging, alerting, and monitoring systems to ensure models effectiveness over time.

• Detect, troubleshoot, and resolve issues such as model drift, degradation, and inefficiencies.

4. Collaboration with Data Scientists & DevOps:

• Work closely with data scientists to ensure that models are production-ready and meet system requirements.

• Collaborate with DevOps teams to integrate MLOps tools and practices into the CI/CD pipeline.

• Optimize model performance by coordinating with various teams to manage the lifecycle of machine learning models.

5. Model Retraining & Continuous Improvement:

• Automate and manage model retraining processes based on incoming new data or changing business needs.

• Create frameworks for evaluating and improving model accuracy, efficiency, and robustness.

6. Security & Compliance:

• Ensure the security of machine learning systems, including data protection, model access control, and sensitive data handling.

• Ensure compliance with relevant regulatory requirements related to data privacy and security.

7. Performance Optimization:

• Work on optimizing models and system performance for faster inference and low-latency predictions.

• Implement techniques like quantization, pruning, and model distillation to optimize the model’s runtime efficiency.

8. Documentation and Reporting:

• Maintain comprehensive documentation for model deployment pipelines, monitoring setups, and operational procedures.

• Provide regular reports on system performance, model health, and operational metrics to stakeholders.

9. Research & Development:

• Stay up-to-date with emerging MLOps technologies and best practices.

• Research and implement new tools and frameworks to improve operational efficiency

Skills and Qualifications:

1. Educational Background:

Bachelor’s or Master’s degree in Computer Science, Engineering, Mathematics, or a related field.

2. Technical Skills:

• Strong experience with cloud platforms (AWS, Google Cloud, Azure).

• Familiarity with Docker, Kubernetes, and containerization technologies

• Proficiency in programming languages such as Python, Java, or Go

• Experience with CI/CD tools (Jenkins, GitLab CI, etc.) and automation frameworks.

• Familiarity with version control systems (e.g., Git).

• Working knowledge of machine learning frameworks (TensorFlow, PyTorch, Scikit-learn, etc.).

• Experience with model serving tools like TensorFlow Serving, TorchServe, or MLFlow.

3. Data Management Skills:

• Strong knowledge of data pipelines and ETL processes.
• Experience with Big Data technologies (Spark, Hadoop, Kafka) is a plus.
• Expertise in data preprocessing and **feature engineering.

4. Monitoring & Logging Tools:

• Proficiency with monitoring tools like Prometheus, Grafana, or Datadog.
• Experience with logging frameworks such as ELK stack (Elasticsearch, Logstash, Kibana) or Splunk.

5. Software Engineering & System Design:

• Strong understanding of **software engineering principles** and best practices.
• Experience in designing highly available, fault-tolerant, and scalable distributed systems.
• Familiarity with **DevOps practices** and agile methodologies.