Senior Machine Learning Engineer- Credit Karma

Responsibilities

• Design, build, and maintain our Next Generation federated ML Platform - built on Vertex AI and Kubernetes.
• Contribute to our python SDK, which enables Data Scientists to efficiently develop, define, and deploy no-human-in-the-loop auto-refreshing deep learning and tree-based ML Models.
• Design, build, and maintain our feature engineering and feature stores services supporting batch and streaming features - built on Vertex featurestore, Chronon, Databricks Tecton.
• Design and build out capabilities supporting training data pipelines and centralized modeling training datasets.
• Provide technical support for owned products, including performing on-call duties, resolving production site issues, and improving the performance and scalability of services.
• Collaborate with cross-functional stakeholders to identify high-impact opportunities, translate business and analytical requirements, develop project plans, and report business value.
• Platform-level monitoring for features, training data, training, and offline batch scoring.
• Provide utilities and capabilities to enable Data Scientists to do pipeline-level monitoring for training and scoring.
• Stay current on innovation trends and propose solutions that integrate those back into our platform.
• Support & mentor other members of our team on current trends, best practices, and their projects.

Qualifications

• MS in Computer Science, Mathematics, Statistics, Machine Learning, or a related quantitative discipline
• 7+ years of industry experience in Machine Learning, Data Science and related areas, ideally in hyper-growth consumer Internet scenarios
• Deep understanding and ability to architect and develop next-generation ML systems, staying ahead of industry trends and integrating latest advancements (e.g. GenAI)
• Strong background in programming languages (e.g., Python, Java, SQL)
• Experience with deep learning frameworks (e.g., Tensorflow, PyTorch)

Preferred Qualifications

• MLOps & Infrastructure: Experience with managing a large-scale platform that services many hundreds of auto-refreshing machine learning models deploying into production with no human-in-the-loop.
• Distributed Processing: Experience with high-volume data processing and frameworks like Spark, Dataflow, Dask, Ray.
• Deep Learning: Extensive Experience with deep learning frameworks such as Tensorflow or PyTorch.
• Google Cloud: Experience with managing platforms backed by Google Cloud ML and AI services.