Products Engineer JD



1. Profile Overview

Strong expertise in managing the full MLOps lifecycle, including model development
support, automated pipeline creation, deployment, monitoring, and continuous
improvement in a scalable, multi-environment setup. They should understand best
practices for model productionization, possess a foundational understanding of machine
learning concepts (types of ML algorithms, evaluation metrics), and collaborate closely
with data scientists. The candidate should also be capable of setting up model monitoring
frameworks to track model quality, data quality, and drift detection.



2. Roles and Responsibilities

- Manage end-to-end ML lifecycle: data ingestion, preprocessing, training, evaluation,
deployment, and monitoring.

- Implement scalable, automated, and reproducible ML workflows across multiple
environments.

- Convert ML models developed by data scientists into production-ready solutions.

- Validate models using appropriate metrics based on problem type.

- Automate ML pipelines using AWS Sagemaker, and event-driven workflows.

- Implement monitoring systems for model performance, drift, latency, and data quality.

- Build and maintain AWS-based ML infrastructure.

- Document pipeline architecture, deployment steps, monitoring setup, and governance.



3. Tech Stack and Corresponding Required Experience

- AWS SageMaker (Model Training, Pipelines, Endpoints, Model Registry): 2 Years

- AWS Services: S3, Lambda, EventBridge, Glue, CloudWatch: 2 Years

- Containerization & Orchestration: ECR, EKS : 2Years

- Airflow for workflow automation : 1 Year



- Python with solid OOPs concepts : 3 Years

- Basic ML understanding: algorithms & evaluation metrics

- Model and data quality monitoring frameworks