Startup: Nexus Labs
After 4 years at Perpetually Innovative Homes (next entry in my portfolio), I founded another startup, Nexus Labs. Working with a three-person team in collaboration with PIHomes, we engineered a prototype smart home system that aimed to significantly reduce energy consumption of residential spaces using various occupancy sensors and machine learning software. I was CEO and Lead Engineer, responsible for brainstorming how to build the system, developing the underlying algorithms that power the system, and building experiments to measure the accuracy of our sensor arrays. Our plan was to eventually create machine learning software that analyzes energy usage across the home, compare it against data on occupancy throughout the home, and predict when were the best times to reduce utility consumption on a room-by-room basis. We also had plans to experiment with the idea of using neural networks that would allow us to train our sensor arrays to recognize human presence in a room while filtering out false positive signals that could be mistaken as a human (I built a prototype in my CS9680 final project - watch a video of me presenting me work here). Towards 2020 Q1, we decided to shutdown Nexus Labs due to the fact that COVID-19 had forced the majority of the workforce to work remotely from home, meaning that, since our smart home system achieved most of its 50% energy reduction by reducing utility usage when people were away from home at work, the value proposition of our smart home system was drastically reduced and we could no longer justify the cost of its development.
“Several years ago, I co-founded a home construction company (Perpetually Innovative Homes, next entry in my portfolio) who’s goal was to build energy efficient houses and, during my time there, I was exposed to a lot of the construction processes. After being exposed to a lot of these processes, I felt that there was a lot of room for improvement in building a more intelligent system to manage the energy usage of a house. So I branched out and started my own company, Nexus Labs, to research this. I discovered that most homes waste tremendous amounts of energy for vast stretches of time throughout the day. So I developed an A.I. powered Energy Management System that reduces a house’s energy usage by up to 50% on weekdays and 20% on weekends - so on weekdays, you’re cutting your heating bill or cooling bill by half: a substantial reduction in energy usage. Its powered by artificial intelligence software and custom sensor arrays, both developed by us and are currently seeking patents. Our target market is residential home builders that build and rent out properties — that includes apartment buildings, condos and sustainable communities such as the West5 development in London, Ontario. We are targeting this market because this industry spends lots of money on any energy-saving technology that they can integrate them into their builds to save on their utility bills. Those savings yield very large profits for them in the long term and, currently, our Energy-Management System is positioned to be one of the best cost saving investments by far. We are seeking funding to hire engineers that will help with the remaining R&D of our system’s development and for scaling up to manufacturing.”
Ryan D (me)
CEO / Lead Engineer
Partner / Civil Engineer
My team consisted of (1) me, (2) a software engineer (Asanka W.) and (3) a colleague from Perpetually Innovative Developments (Samy, who still worked there and decided to contribute his spare time to Nexus Labs). He is a civil engineer and understood the ins and outs of building residential living spaces. As such, his knowledge helped guide us on how to develop and integrate our energy management system into common residential building schematics. Myself and Asanka were tasked with developing the prototype software and hardware to be used as a proof-of-concept product that could then be showed off to investors. I developed various algorithms that analyzed data from our sensor arrays and devised experiments to help test the accuracy of our sensory arrays. Asanka then integrated my algorithms with the rest of the system’s software and hardware. Samy was intermittently consulted throughout development.
A detailed proposal was created for our website, www.nexusone.ai. It covers how we thought was the best way to educate potential buyers of the system. The proposal was also used to give investors and other stakeholders an overview of how the technology works.