Data. Innovation. Insurance? You bet!
As a trained meteorologist, I never imagined that I would end up working for an insurance company. And yet here I am, four and a half years later, working for a team that started with eight employees and now has 80, with the objective to almost double in size by the end of the year.
This team is called the Intact Data Lab. The arrival of artificial intelligence has revolutionized the insurance industry and presented me with the opportunity to work on many stimulating projects related to weather impacts.
Here are some of the interesting projects we’ve worked on so far:
Weather Impacts
Climate change isn’t just happening tomorrow: it’s happening today. In order to understand those changes, we first need to understand how weather impact our results. Feeding weather, climate and topographic data into machine learning models, we developed risk maps that assess the risks of Canadians to weather perils such as heavy rain, hail and wildfire. We can therefore know which regions are more vulnerable.
To mitigate potential losses for our commercial line of business, we have developed a forecast system that automatically detects which policies are at risk of roof collapse due to the weight of snow. This allows us to provide subsidies to our clients to pay for the snow removal and prevent damage.
Telematics
Intact collects more than 20 data points per second from each of our voluntary customers via the Intact app — my Driving Discount. We are then able to model driving habits, such as how much you use your phone while driving, acceleration and braking patterns, etc. We then give our customers discounts based on how safe their driving behaviour is.
To determine what is considered safe and unsafe driving, the Data Lab analyzed billions of data points from hundreds of million of trips and built AI solutions that capture those driving events. We are currently trying to understand how the environment around a driver — such as weather — impacts driving behaviour.
Fraud Detection
Insurance is a business based on trust, both from the insurer and the client. Unfortunately, some clients will stage scenarios that will allow them to get paid from the insurer when they shouldn’t. The Data Lab built a solution to detect fraud by building AI tools that can detect fraud from unusual patterns in the claim process.
And those are just a few examples of what we’re doing right now. I’m proud to work on such innovative projects, surrounded by so many talented people. Looking forward for our next projects… and colleagues!
Want to work with me? Check out these opportunities.