Building Intelligent Energy Systems
WattsIQ was founded in 2019 by a team of engineers, data scientists, and energy specialists committed to making electricity management smarter and more sustainable through applied artificial intelligence.
WattsIQ was founded in 2019 by a team of engineers, data scientists, and energy specialists committed to making electricity management smarter and more sustainable through applied artificial intelligence.
From research lab to deployed systems optimizing energy across Canada.
WattsIQ emerged from research conducted at McGill University's Department of Electrical and Computer Engineering, where predictive models demonstrated significant potential for reducing building energy consumption. The founding team recognized that while utility-scale grid optimization had received substantial attention, individual buildings and facilities lacked access to sophisticated AI-driven management tools.
The company's initial focus was residential energy management in Quebec, where hydroelectric power dominance and time-of-use pricing created opportunities for load shifting. Early deployments validated the technology's effectiveness and identified requirements for commercial and industrial applications. WattsIQ expanded its platform to support diverse building types, operational contexts, and regulatory environments.
Today, WattsIQ systems operate across Quebec and Ontario, managing electricity consumption for homes, offices, manufacturing facilities, and institutional buildings. The company continues to invest in research and development, refining algorithms and expanding integration capabilities.
A Canada where energy intelligence becomes part of everyday life. Buildings, homes, and facilities operate with the efficiency of well-tuned systems, minimizing waste and environmental impact without compromising function or comfort.
To make electricity management smarter and more sustainable by offering real-time insights and predictive recommendations that align efficiency with environmental responsibility through applied artificial intelligence.
The values that guide our engineering decisions and business practices.
Every algorithm is validated against performance benchmarks. Models are tested on diverse datasets before deployment. Continuous monitoring ensures maintained accuracy.
Users receive explanations for system recommendations. Model decisions are interpretable through SHAP values and feature importance rankings. No black-box operations.
Energy consumption data belongs to customers. Aggregated analytics never expose individual patterns. PIPEDA compliance guides all data handling practices.
Carbon reduction is measured and reported. Systems prioritize low-emission grid periods. Renewable integration is optimized for maximum impact.
Solutions work within existing infrastructure. Deployments minimize disruption. Training ensures successful adoption by facility teams.
Models evolve with new data and techniques. User feedback informs product development. Research partnerships advance the state of the art.
Multidisciplinary expertise in machine learning, electrical engineering, and building systems.
PhD and Masters-level researchers specializing in time-series forecasting, reinforcement learning, and anomaly detection. Backgrounds in computer science, statistics, and computational engineering.
Licensed professional engineers with expertise in electrical systems, HVAC design, and building automation. Experience with BACnet, Modbus, and industrial control protocols.
Full-stack developers and DevOps engineers maintaining cloud infrastructure, data pipelines, and user interfaces. Expertise in Python, React, and distributed systems.
Installation technicians, commissioning specialists, and customer support engineers ensuring successful deployments and ongoing system performance.
Current areas of investigation advancing energy management capabilities.
Training models across distributed installations without centralized data collection. Privacy-preserving algorithms that improve with scale while protecting individual consumption patterns.
Accelerating model training for new installations by leveraging patterns from existing deployments. Reducing commissioning time and improving initial performance.
Coordinating multiple reinforcement learning agents to optimize complex facilities with interdependent systems. Hierarchical control strategies for large-scale environments.
Advancing interpretability techniques for energy models. Generating natural language explanations for system recommendations and decisions.
Deploying inference models directly on local hardware to reduce latency and enable operation during connectivity outages. Balancing edge and cloud processing.
Protocols for deeper integration with utility systems. Real-time carbon intensity optimization and automated participation in grid services markets.
WattsIQ collaborates with research institutions to advance energy management methodologies and validate new approaches.
Department of Electrical and Computer Engineering - Joint research on predictive control algorithms and grid integration.
Institute for Sustainable Energy - Collaboration on renewable integration and microgrid optimization.
Advisory role in federal energy efficiency programs and building performance standard development.
Our commitment to responsible AI deployment in energy systems.
Every automated decision can be traced to specific input features and model logic. Users receive plain-language explanations for why the system recommends particular actions. SHAP values quantify the contribution of weather, occupancy, pricing, and other factors to each prediction.
Automation serves users, not the reverse. Manual overrides are always available. Preference settings allow customization of comfort ranges, equipment behavior, and optimization priorities. The system adapts to user feedback.
Energy consumption data remains the property of customers. Aggregated analytics for model training never expose individual patterns. Data deletion requests are honored. Third-party data sharing requires explicit consent.
Models are trained on diverse datasets representing various building types, climates, and operational patterns. Performance is validated across demographic and geographic segments to ensure equitable outcomes. Bias audits are conducted regularly.
WattsIQ is growing. We seek engineers, researchers, and specialists passionate about applying technology to environmental challenges.