ML OPS

TechDigital Group; San Jose (CA); On-site; USD 140,000 - 190,000; Full time

Job summary

TechDigital Group is seeking an experienced MLOps Engineer to design, deploy, and maintain scalable ML systems in production. You will own the full lifecycle of ML pipelines from data ingestion to model serving.

Responsibilities include building data and feature pipelines on Kubeflow and Vertex AI, containerizing models with Docker, implementing CI/CD for ML workflows, and monitoring data quality and experiment tracking in a production environment.

Qualifications

• Experience building end-to-end ML pipelines and data workflows.
• Hands-on with Kubeflow, Vertex AI Pipelines, and Google Cloud Platform.
• Containerization with Docker and deployment of ML models.
• CI/CD practices for ML workflows and automation.
• Experience with TensorFlow and ML experiment tracking.

Responsibilities

• Design, build, and manage automated data ingestion, transformation, and validation pipelines.
• Containerize ML models and dependencies with Docker; manage images.
• Implement and monitor CI/CD pipelines for ML models and datasets.
• Define and run hyperparameter tuning and experiment tracking workflows.
• Deploy real-time inference endpoints and maintain scalable serving stacks.

Skills

ML OPS

Docker

Kubeflow

Vertex AI

TensorFlow

CI/CD

Tools

Kubeflow

Vertex AI

Docker

Artifact Registry

Cloud Storage

TensorFlow

Job description

Mandatory Skills:

ML OPS

JD:

We are looking for a skilled MLOps Engineer to join our team and help us build, deploy, and maintain robust and scalable machine learning systems. You will be responsible for the full lifecycle of our ML pipelines, from data ingestion to model serving. This is a hands-on role where you will design and implement automated workflows, ensure data quality, and manage model deployments in a production environment.

Responsibilities

Data and Feature Pipelines

Design, build, and manage automated data ingestion, transformation, and validation pipelines using services like Kubeflow Pipelines and Vertex AI Pipelines.

Feature Engineering

Implement and containerize feature engineering logic for diverse datasets, ensuring reusability and scalability.

Data Validation

Integrate and manage data validation processes, including leveraging advanced techniques like AI Agents and the Generative Language API to automatically detect and remediate data quality issues.

Model Training and Experimentation

• Set up and maintain automated continuous training (CT) pipelines using Vertex AI Pipelines (Schedules) and Cloud Scheduler.
• Implement experiment tracking to log and compare model parameters, metrics, and artifacts.
• Configure and execute Hyperparameter Tuning jobs using Vertex AI Training to optimize model performance.

Deployment and Serving

• Containerize ML models and their dependencies using Docker and manage images with Artifact Registry.
• Build and maintain CI/CD workflows for ML models, ensuring seamless and automated deployment.
• Configure and manage low-latency production serving environments using Vertex AI Endpoints for real-time inference.

Qualifications

• Strong experience with Google Cloud Platform (GCP) services, specifically in the MLOps and ML domain (Vertex AI, Kubeflow, Cloud Storage, Artifact Registry).
• Proven ability to design and implement end-to-end ML pipelines for data management, model training, and deployment.
• Hands-on experience with containerization technologies like Docker.
• Familiarity with CI/CD practices and pipeline automation.
• Knowledge of ML frameworks like TensorFlow, and experience with experiment tracking and hyperparameter tuning.
• Excellent problem-solving skills and a strong understanding of the ML lifecycle.
• Experience with the Generative Language API (Gemini model) or other AI Agent integrations is a plus.