Physical AI Engineer (Control, Optimization, Validation)

About the role

PassiveLogic, the first fully autonomous platform for buildings, is looking for a Physical AI Engineer to develop, optimize and validate its autonomous control system, AI models and levels of autonomy. The engineer builds algorithms that let buildings understand their physical state, learn from experience and autonomously optimize their operation, combining control theory, optimization, machine learning, physics-based simulation and systems validation across digital twins, autonomous reasoning, building physics, equipment behavior and human comfort.

Key responsibilities

• Design and implement autonomous control strategies using physics-based digital twins, translating comfort, energy, equipment life and operational needs into control objectives, constraints and cost functions.
• Develop scalable optimization algorithms (stochastic gradient descent, coordinate descent, distributed optimization, Bayesian methods, evolutionary algorithms) and physics-based fault detection and diagnosis algorithms.
• Create fault-tolerant control methods and integrate control and learning algorithms into production systems with reliability and real-time performance.
• Build physics-informed predictive and learning models using deep learning, reinforcement learning and transfer learning, including autonomous agents, state estimation, model adaptation and control-correction algorithms.
• Validate systems by defining performance metrics, acceptance criteria and automated tests, and comparing simulation results with analytical solutions, lab measurements and real-building data.

Required profile

• MS or PhD in control engineering, computer science, robotics, applied mathematics, mechanical engineering or a related field.
• Demonstrated expertise in AI development, scientific machine learning, optimal control, reinforcement learning, multi-agent systems, physics-informed machine learning and optimization theory.
• Strong background in control theory, model predictive control, state estimation and system identification.
• Strong analytical, debugging, root-cause analysis, communication and collaboration skills.

Required skills

• Strong programming in Python, C++, Swift or a similar language.
• Nice to have: automatic differentiation and differentiable programming; software design patterns and architecture; building science, HVAC, thermodynamics, energy modeling or grid-interactive controls.
• Nice to have: software-in-the-loop and hardware-in-the-loop testing, formal methods, probabilistic modeling, graph neural networks, Dymola and EnergyPlus, vector, SIMD and tensor computation.

What we offer

• Competitive compensation and a generous equity package.
• Medical, dental and vision coverage, disability and life insurance options.
• Flexible paid time off, team-building events, free catered lunch Monday through Friday, free National Parks pass, free ski pass and free ClassPass access.

Candidates should apply with a cover letter, a resume and a description of a personal project they created or designed.