About the role
A cross-functional expert who bridges Data Science and Software Engineering skill sets to build scalable, production-ready AI solutions. Builds re-usable feature tables, ETL pipelines, and optimises model training and inference for performance. Builds monitoring systems to enable rigorous tracking of ML-powered solutions in production.
Responsibilities:
- Data Science, Product, and Engineering Stakeholders: You will work closely with the Data Science, Product, and Engineering stakeholders to build reliable, scalable and optimised ML pipelines, ensuring seamless deployment, monitoring of the performance of ML solutions, and maintainability standards.
- Programmatic Traders, Analysts, Sales teams, Account Managers: ML solutions in production will be consumed as insights, recommendations, and campaign optimisations by personas that plan and manage programmatic campaigns and insights.
Requirements:
- 2 - 4 years of hands-on experience deploying and monitoring Machine Learning Models in production.
- Strong proficiency in Python and familiarity with distributed data processing frameworks like PySpark.
- Excellent grasp of the basics of Machine Learning, Statistics and Probability.
- Experience with building scalable ETL pipelines for efficient storage, processing and retrieval of features to support ML model development.
- Experience building and deploying in production end-to-end model training, inferencing and retraining pipelines.
- Experience with data and model monitoring to ensure consistency and reliability of solutions in production.
- Familiarity with standard best practices like version control and CI / CD for code management, unit testing and quality code standards.
- Experience working in cross-functional teams, consisting of Data Scientists, Data Engineers and Product Managers to deliver quality solutions that meet business goals.
- Passion for self-driven learning and an eagerness to keep up with continuously evolving MLOps / LLMOps technologies and best practices.
- Hands-on experience with distributed data processing frameworks like Ray.
- Hands-on experience with designing and maintaining Feature Stores.