About the Role
About Infinite Cooling
Born at MIT, Infinite Cooling is eliminating cooling bottlenecks across critical industries: in data centers, we unlock compute capacity by reducing cooling power needs. In power plants and industrial facilities, we increase output by lowering process temperatures.
Our AI-driven software platform, TowerPulse™, uses proprietary Multiphysics AI algorithms trained on large sets of real customer data to continuously monitor and optimize cooling tower performance, boosting cooling capacity while reducing energy, water, and maintenance costs.
We are a venture-backed startup based in the Boston area, with over $28M raised. We have a growing base of customers and proven deployments, and we are now scaling our commercial team to accelerate growth.
About the Role
We are looking for a Research Scientist, Applied ML, to bring statistical and mathematical depth to our team. You will work at the seam between first-principles models and data: deciding where physics should constrain a model, and where learning & statistical methods should fill the gaps. You will need to be rigorous, curious and comfortable taking a model from notebook to production with our software team. This is applied research: you will read and test new methods, and success is measured by what reaches TowerPulse™ and our customers.
What you will do
Build ML models that combine our first-principles models with learned components and stay accurate with limited site data
Develop probabilistic forecasting and anomaly detection on multivariate sensor time series to catch anomalies in both customer equipment performance & instrumentation health
Use transfer learning, hierarchical models and time-series foundation models to onboard new sites in weeks rather than months, and benchmark adopted models against our own
Own the model lifecycle with our software team: evaluation harnesses, backtests, drift monitoring and retraining triggers running in our AWS streaming and batch pipeline
Turn model output into clear engineering findings for plant operators, customers and our commercial team
Requirements
MS or PhD in statistics, applied mathematics, machine learning, computer science or a related quantitative field, or equivalent experience
Experience building models on real-world time-series or sensor data, with at least one model used in production or by real decision-makers
Depth in statistical modeling and the judgment to know when a simple model is the right answer
Fluency with AI-assisted development tools, while owning the correctness of what you ship
Clear writing and speaking, including explaining uncertainty to people who are not statisticians
Contact
Share this position