MLops Tech Lead

Hands On experience in DevOps and MLOps practices, with a focus on managing cloud-based machine learning environments.
Model Deployment Build, optimize, and maintain cloud based environments for deploying, monitoring, and scaling machine learning models and data pipelines.
Package machine learning models into Docker containers (Relative experience in ML models)
Solid foundational knowledge of Azure Open AI and GPT LLM model fine-tuning techniques, with a strong grasp of prompt engineering principles.
Develop and automate the unified CI/CD pipelines in Azure DevOps
Hands-on experience in containerization and orchestration tools such as Docker and AKS
Work closely with data scientists to ensure smooth handoffs and integration of machine learning models into production systems.
Automate model testing, validation, and performance monitoring for containerized solutions.
Deploy and manage code using Azure Repos, and Azure DevOps (CI/CD, Docker, Function Apps), and utilize Kubernetes Helm charts for model deployment. Implement Docker best practices.
Design and implemented a complete MLOps pipeline utilizing Azure Machine Learning in conjunction with open-source frameworks
Dockerize ML model training and serving processes, containerizing them according to specific versions, and then deployed the containers to an Azure Kubernetes environment.
Solid foundational knowledge of Azure Open AI and GPT LLM model fine-tuning techniques, with a strong grasp of prompt engineering principles.
Contribute to the establishment of a unified CI pipeline, facilitating the efficient use, synchronization, and application of common templates and files across downstream repositories using Copier and Azure Repos.
Enforce best coding practices within the MLOps pipeline, including linting, unit testing, and version validation.
Set up and deploy the MLflow stack for model experiment tracking, version management, and registry.
Implement Docker best practices, optimizing Dockerfile and Docker Compose to minimize Docker image size.
Configure data drift, target drift, and data quality metrics with Evidently and developed a user-friendly web app for easy drift detection between training and production data.
Deploy and manage code using Azure Repos, and Azure DevOps (CI/CD, Docker, Function Apps), and utilize Kubernetes Helm charts for model deployment.
Work closely with the cross-functional team including data science team, engineers etc.
Document deployment workflows, best practices, and system configurations for reoducibility.
Experience in AKS Cluster setup.
Experience in cloud-native tools for monitoring containerized application, auto-scaling and load balancing.
strong understanding on machine learning lifecycle and model integrations.
Tools and techniques to have hands on experience.
Python
Machine Learning Frameworks
Docker
AKS
Azure ML
Linux/Shell scripting
Data and Model Drift monitoring (EvidentlyAI)
Kubeflow
Azure DevOps, AutoML
MLFlow