Senior ML Engineer, Building Energy Optimization

The challenge

The current control models are trained per-building and take 6 weeks of data to calibrate. You will develop a transfer learning approach that gets to 80% of optimal performance within 48 hours of installation.

Role overview

ML models controlling HVAC systems in commercial buildings. Weather forecasts, occupancy sensors, zone temperatures, and energy prices go in. Setpoints that minimize energy consumption while maintaining comfort come out. When the model drifts, buildings get uncomfortable and the phone rings.

Specializations

Technical requirements

• MS in Machine Learning, Controls, or Mechanical Engineering with ML focus
• 5+ years applying ML to physical systems (buildings, energy, HVAC, or industrial processes)
• Experience with reinforcement learning applied to real-world control problems
• Proficiency in Python, PyTorch, and time-series data processing
• Understanding of HVAC system physics (psychrometrics, heat transfer, fluid flow)
• Experience deploying ML models with safety constraints (bounded outputs, fallback controllers)

Key responsibilities

• Develop reinforcement learning and model predictive control algorithms for HVAC optimization
• Design transfer learning architectures that accelerate model calibration for new buildings
• Build feature engineering pipelines that process BMS, weather, and occupancy data into model inputs
• Validate model performance against baseline energy consumption using A/B testing frameworks
• Collaborate with controls engineers on safe deployment of ML-generated setpoints to building equipment
• Monitor deployed model performance across hundreds of buildings and diagnose drift or degradation

Compensation & Benefits

• Base salary: $170,000 - $215,000
• Equity (stock options, 4-year vest)
• Annual bonus 10-15%
• Fully remote with co-working stipend
• 4 weeks PTO