Assembled is a company that provides a unified platform for customer support, balancing human agents and AI at scale. The role focuses on developing forecasting interfaces, optimizing scheduling for support agents, and enhancing MLOps for efficient machine learning operations.
Responsibilities:
- Predicting contact volume: Developing forecasting interfaces, data pipelines, and inference servers to predict support contact volume and determine the optimal number of support agents required for specific days and times
- Scheduling 1000s of support agents: Designing and implementing interfaces to collect and store team preferences and customer business constraints (e.g., labor laws), enabling the creation of optimal schedules for teams of thousands of support agents based on these forecasts and constraints. (check out https://en.wikipedia.org/wiki/Nurse_scheduling_problem)
- MLOps: Enhancing machine learning efficiency and operations to support rapid model deployment and iteration
Requirements:
- Experience using Python libraries like pandas, SciPy, and seaborn for statistical or predictive work
- Previous experience working on a machine learning or algorithmic team
- A strong commitment to advancing both statistical and runtime performance, ensuring reliable and efficient forecasting and scheduling