ML Ops Engineer

• Develop and enhance the ML platform to standardize model development and deployment workflows.
• Create reusable components to streamline the Data Science teams efforts and expedite the model lifecycle.
• Integrate models seamlessly with various products and systems.
• Implement robust logging and instrumentation for monitoring scoring requests for models in production.
• Establish systems for continuous model monitoring and trigger mechanisms for retraining based on performance metrics.
• Design and build A/B testing frameworks with support for canary deployments and shadow models to evaluate different model versions.
• Integrate data pipelines necessary for model retraining and update activities in production.
• Scale training and inference capabilities using standardized environment setups and deployment strategies.
• Incorporate open-source frameworks and proprietary tools into the MLOps pipeline to achieve development goals.
• Prototype and evaluate different open-source frameworks to identify optimal technology stacks for the pipeline.
• Focus on CI/CD pipeline integration for models and ensure seamless deployments in various environments.

Experience Expertise :

1. 3+ years of experience in MLOps, software development, and deploying machine learning models.
2. Strong programming skills in Python, with experience in scripting languages for automation.
3. Must have extensive experience with Docker and Kubernetes, including the ability to create, manage, and debug containerized environments, as we'll as deploy and monitor Data Science solutions in production.
4. Excellent problem-solving abilities and analytical mindset.
5. Experience with deploying and monitoring Data Science solutions in production environments.
6. Hands-on experience with MLOps tools and frameworks such as MLFlow, Seldon, KubeFlow, and cloud services like AWS Sagemaker, Azure ML, or GCP AI Platform.
7. In-depth understanding of public cloud infrastructure and services (AWS, Azure, GCP).
8. Proven track record of evaluating and building prototypes with open-source tools to guide technology choices.