ML Ops-Model Deployment & Monitoring

Pineswift Technologies; Gurugram District; On-site; INR 1,800,000 - 2,800,000; Full time

Job summary

Pineswift Technologies in Gurugram, India seeks an ML Ops engineer to partner with data scientists to package models for production and deploy them for batch or real-time scoring.

You will build production scoring pipelines using Python and PySpark, manage model artifacts, monitor drift and performance, and collaborate with governance and security teams to meet enterprise requirements.

Qualifications

• Strong hands-on proficiency in Python and PySpark.
• Experience deploying ML models to production.
• Knowledge of model evaluation metrics, drift and performance interpretation.
• Experience building production scoring pipelines and monitoring operations.
• Familiarity with model versioning and lifecycle-management tools.
• Experience with CI/CD, automated testing and dependency management.
• Troubleshooting across data pipelines, model execution and deployment environments.
• Working knowledge of banking domain and governance for production models.

Responsibilities

• Partner with data scientists to package models for production deployment.
• Deploy models for batch scoring or real-time inference.
• Develop production scoring pipelines using Python and PySpark.
• Automate deployment validation, versioning and release workflows.
• Manage model artifacts, dependencies, configurations and deployment history.
• Monitor model performance, data drift, scoring failures and reliability.
• Implement alerts, troubleshooting procedures, rollback mechanisms and recovery processes.
• Support model refreshes and redeployments with model owners.
• Collaborate with technology, security and governance teams.

Skills

Python

PySpark

Model deployment

CI/CD

Data pipelines

Monitoring

Governance

Azure/Databricks

Job description

Greetings

We have a few requirements for ML Ops. PFB the details:

Key Responsibilities

• Partner with data scientists to package models and prepare them for production deployment
• Deploy models for batch scoring or real-time inference based on business requirements
• Develop production scoring and supporting data pipelines using Python and PySpark
• Automate deployment validation, versioning and release workflows
• Manage model artifacts, dependencies, configurations and deployment history
• Monitor model performance, data drift, scoring failures and operational reliability
• Implement alerts, troubleshooting procedures, rollback mechanisms and recovery processes
• Support approved model refreshes and redeployment in collaboration with model owners
• Work with technology, security and model-governance teams to meet enterprise requirements
• Maintain deployment documentation, monitoring controls and operational runbooks

Required Skills & Experience

• Strong hands-on proficiency in Python and PySpark
• Practical experience operationalizing and deploying machine-learning models.
• Sound modeling knowledge, including evaluation metrics, model behaviour, drift and performance interpretation
• Experience building production scoring pipelines and monitoring model operations
• Familiarity with model versioning and lifecycle-management tools
• Experience with CI/CD, automated testing and dependency management
• Ability to troubleshoot issues across data pipelines, model execution and deployment environments
• Working knowledge of the banking domain and the governance requirements associated with production models

Preferred Skills

• Experience with Databricks and Azure-based deployment environments
• Familiarity with APIs, containers and orchestration platforms
• Experience supporting model approvals, audit documentation and controlled releases