stub Kaitlyn Albertoli Founder of Buzz Solutions - Interview Series - Unite.AI
Connect with us


Kaitlyn Albertoli Founder of Buzz Solutions – Interview Series

Updated on

Kaitlyn Albertoli is the founder of Buzz Solutions, an AI company that provides asset fault detection and predictive analytics for powerline inspections, providing critical savings in preventing downed lines, power outages, and sparked wildfires due to failed grid infrastructure.

Buzz Solutions was founded as part of the Stanford Launchpad course in 2017, could you share some details regarding these early days?

We started Buzz Solutions to address a critical need that arose during the early days of power utility infrastructure inspections. During this pivotal time, utility companies began to collect more visual data to ensure and enable thorough and frequent inspections. From the outset, we spent a great deal of time delving into the pain points of utilities, as well as understanding the near and long-term vision for their inspection programs. 

Recognizing that each utility has its own unique means of inspection and routine process, it became clear that the best entry point into the market was through highly- accurate and flexible AI-powered algorithms. In the first two years, our mission was to build the most accurate and easily retrainable algorithms in the market that utility companies could deploy directly into their existing systems. We launched the official Buzz Solutions PowerAI product into the utility market in August 2019.

Utilities are mandated to inspect all transmission and distribution infrastructure, why is this such a problem for the legacy manual inspection methods? 

As utilities are mandated to conduct more frequent inspections, data collection levels are skyrocketing. Utility companies are collecting 5-10x the data of historical levels, oftentimes collecting hundreds of thousands and millions of images annually. The current analysis process of this data is done manually, with linemen and field technicians, which is a highly-tedious and unscalable process. As inspections become more frequent, the manual process becomes more expensive, time intensive, and lends to increased risk of infrastructure failure due to data not being processed in a timely manner. 

What type of visual processing data is captured in the field?

Images and video streams are currently captured in the field using drones, helicopters, fixed-wing aircrafts, and even ground-based data capture. Drones are becoming a more prevalent means of inspection because they can fly closer to structures and collect images from various angles that aren’t possible with manned aircraft. Drones are able to capture visual imagery of various electrical components, power grid structures, surrounding vegetation and locations. This enables a more comprehensive inspection, so a utility can better understand the health of each infrastructure component for both T&D power lines and substations. 

What type of cost savings are seen by analyzing these images with AI versus manual analysis?

Analyzing imagery with AI provides tremendous cost savings, which continue to increase over time. AI provides a direct, initial cost savings of about 50% compared to manual analysis, and with time, those savings increase exponentially as the AI tracks trends and gets smarter over time. This enables more targeted, informed inspections and provides linemen additional savings by delivering better information so they can more clearly and quickly plan a path to maintenance. 

The Buzz Solutions technology can identify what needs to be fixed in just a few hours, could you discuss the AI that is used to enable this?

PowerAI Machine Vision algorithms are trained to detect a specific list of anomalies for utility infrastructure. We spent two years building these algorithms from scratch and aggregating varied datasets across geographies and timelines to train the AI to encompass these faults. An advantage we have is that we trained our AI with real images vs. “synthetic” images and our accuracy of identifying and predicting equipment faults or issues is significantly higher than the industry average. This means that utilities can fix issues much more quickly and efficiently. 

Additionally, our AI leverages human-in-the-loop training, where the field technicians and engineers feed data back into the AI enabling the model to get smarter and more personalized over time. The comprehensive list of failure modes that the PowerAI algorithms detect today have been derived from the biggest needs that utilities have expressed.

Could you discuss the predictive analysis system that is used and the benefits that it offers?

Buzz tracks utility asset trends and failures over time, ultimately helping the AI and Machine learning systems become stronger, more personalized, and more efficient. This also pushes the systems to derive insights from these trends and predict areas which may be prone to potentially higher areas of fault i.e. “Hotspots”. This is where the true potential of a predictive analytics system comes into play and enables utilities to have better insight into where and when their equipment may fail. 

Could you discuss your plans to also target the wind and solar sector?

To date, Buzz has focused on becoming the most accurate and effective AI solution in the utility inspection space. That being said, there are many other areas of infrastructure including renewable energy generation where inspection analytics are needed and are quite valuable. Buzz has plans to expand beyond the T&D inspection space, and will make announcements on some of those more concrete market expansions as there are new use cases which we add into our portfolio. 

How does optimizing the power sector assist with climate change?

Buzz Solutions aids in sustainability-focused efforts and assists with some of the biggest climate-related issues we face today by enabling reduced grid-induced disasters, reduced emissions, and increased grid reliability. Our AI-based fault detections reduce wildfires sparked by faulty assets as we alert utilities to failures and vegetation encroaching on the infrastructure. 

Additionally, our systems flag common fault areas (“hot spots”). Pre-determined hot spot areas enable targeted inspections rather than helicopters aimlessly flying for hundreds of miles. Targeted inspections help utilities reduce carbon emissions and foster predictive responses instead of reactionary actions. Our technology enables a more resilient and stable grid, allowing efficient penetration of renewable energy resources on the grid infrastructure.

Could you discuss your overall vision of digital transformation of the utility sector?

Buzz Solutions is at the forefront of the digital transformation of the inspection and maintenance workflow for power utilities. While the collection of more data is important, it is even more significant to successfully manage the data and derive actionable insights from that information. This is where Buzz is particularly valuable. 

Not only does our solution PowerAI provide fast insights into the current health of infrastructure, it also tracks this data and alerts a utility to an area that poses the most potential risk to the grid. PowerAI allows for faster upgrading of components and a more efficient path toward grid modernization. Digital transformation in the sector has enabled seamless data collection, but the power of the data is being able to turn the raw data into a cohesive picture and derive specific insights from that information. 

Thank you for the great interview, readers who wish to learn more should visit Buzz Solutions.

A founding partner of unite.AI & a member of the Forbes Technology Council, Antoine is a futurist who is passionate about the future of AI & robotics.

He is also the Founder of, a website that focuses on investing in disruptive technology.