The mission: safer driving experiences. The engine: data.


Even with decades of industry experience, Min Shao regularly learns new things as VP of Analytics at Arity.

“I’ve been working in data and data analytics for almost 30 years. I thought I already knew a lot,” Shao said. “But every day I learn how insurance works and how we can use data in the right way to solve insurance and transportation issues.”

Shao leads a team of around 50 scientists and data analysts who seek to further Arity’s mission – an ambitious mission.

We want to make transport smarter, safer and more useful for everyone. To do this, we collect and analyze huge amounts of driving data to understand how people move,” Shao said. “We use the insights we get from data analytics to help deliver better experiences on and off the road – for drivers as well as communities and businesses.”

By bringing together different types of incoming data, Arity aims to unlock critical and beneficial information for auto insurance companies and drivers. According to the director of actuarial and data governance, Megan Klein, this is a powerful equation.

“We can take driving data and combine it with insurance results data. By combining it, we understand not only how people move, but how how they move can affect their risk profile,” Klein said, “That brings us to the part of the mission statement where we help make transportation smarter, safer and more useful. When people understand how they move and how that can impact their risk of potentially having an accident, we can really coach them and encourage them to take advantage of this data and make the road safer.

The scale of these data? Significant, to say the least. Leading a team of engineers who help ensure Arity’s platform is up-to-date and data pipelines are running smoothly, Senior Director of Engineering Rahul Chandrawanshi has compared to “a sea of ​​data entering our system”.

“The data continues to grow as we onboard more customers, get data from different sources, and help more drivers,” said Chandrawanshi, who joined Arity five years ago directly from his founding company, The Allstate Corporation. .

All of this data requires more than technology to manage it, and Arity has the human component broken down. If a new technology stack and platform provides the engine to achieve Arity’s goals, then it’s thoughtful teamwork between different departments that helps manage the complexity and nuances of their roles in what Shao has called “one of the most collaborative business types I’ve ever seen”. works at.”

“Megan’s Data Governance and Actuarial and Rating Services team, as well as the Operational Excellence team, are very helpful in defining things like: what data we should ingest and what data we shouldn’t ingest. How do we protect the privacy of our customers How do we meet regulatory requirements?” Shao said.

Below, Klein, Shao and Chandrawanshi explained why their roles translate into stimulating technical challenges and how, guided by Arity’s mission, they are able to make a difference.

Team members walking past a car in the office with the Arity logo on it

Let’s talk about delivering safe experiences. How does this translate into work?

Director of Data Governance and Actuarial and Rating Services Megan Klein: We offer our own mobile app, called Routely, as well as software that can be integrated with an insurance company’s mobile app. By comparing telematics data on driving behavior with complaints information, we are able to create insights that can help insurance companies better understand driver behaviors and change or adjust the rates they charge based on those behaviors. This acts as an incentive for drivers, who now understand that they should benefit from a lower rate if, for example, they never touch their mobile phone while driving. This makes them safer drivers.

Vice President of Analytics Min Shao: We integrate our crash detection machine learning model into our software development kit. If you have installed our partners’ apps with our built-in SDK, the model will automatically detect if there is a crash that just happened in real time. Then it automatically sends alert messages to emergency contacts and service providers, as well as your family and friends. It’s another example of improving the safety and usefulness of the driver experience.

Help make insurance premiums fair

When it comes to applying a data-driven philosophy to his work, Arity focuses on the information he finds most relevant: driving behavior. This, in turn, helps to make the practice of pricing insurance premiums more accurate. “Compared to some of the traditional factors that insurance companies use to price their premiums, such as gender or age, tying driving behavior is much fairer,” Shao said.

What is the scale of the data your teams work with respectively?

Rahul Chandrawanshi, Senior Engineering Manager: The scale of the data is huge. We process nearly 60 million trips a day. We have collected hundreds of billions of kilometers of driving data and collected nearly 700 million kilometers of driving data every day. The data is in RAW format. We ingest data, process data, normalize data, anonymize data and make it usable for customers. This process happens in our pipelines, where we use a different set of tools and technologies to make everything happen in near real time.

The scale of the data is huge. We handle nearly 60 million journeys a day.

Klein: It is also this scale of data combined with other data sources – which includes, as mentioned, a lot of information on the results of insurance claims. By having this large scale of driving behavior that is very well managed by our engineering team, and combining that with insurance information, it’s this scale that no other company really has access to. We can create truly meaningful ideas.

How do you ingest the data you work with?

Chandrawanshi: We get data from mobile devices, telematics devices and sometimes from the car itself. Our infrastructure is built on the cloud. We ensure that all data that comes in is authorized by organizations we trust and is already registered with Arity.

We use many advanced tools and technologies within the cloud infrastructure to ensure that we are able to handle so much data. Because we collect a lot of traffic data, there are peaks and troughs in data that occur during any 24-hour period. We have enabled autoscaling on our endpoints, which allows us to grow and contract based on the amount of incoming data.

When the data reaches our terminals, we perform validation and access authorization. After that, we use different streaming tools, like Kafka, Flink, NiFi, etc. and from there it is kept in our operational and analytical data stores, which is a group of databases like Cassandra, Postgres, EMR, etc. Most of our architecture is an event-driven architecture, which means we have the data flowing through our pipelines which is used by different products depending on business needs.

chao: Pretty much anything you can think of, we’ve probably tried. We use a diverse set of tools; there aren’t really any specific requirements to use certain tools. This gives analysts and data scientists the opportunity to experiment and use the best tools they know so that they feel most comfortable with their work.

Pretty much anything you can think of, we’ve probably tried. We use a diverse set of tools.

Arity team members working in the office

What is an exciting project you are working on at the moment? What is the impact of this project and how does it benefit Arity customers?

chao: We look at how we can actually deploy machine learning models to the edge without having to release software releases. With this we can change the templates without the user updating their apps. This is very exciting because it gives us a lot of flexibility to quickly improve the performance of the model.

Chandrawanshi: We are moving towards managed services where our technology partners will take care of infrastructure maintenance, giving Arity engineers more time to work on building code and products. It reduces operational overhead and increases availability for our end customers by making the system more robust and performing better.

Klein: In the field of insurance, it sometimes happens that information on the behavior of the driver is only partially available. What is the impact of only having information on a subset of drivers or vehicles? Our team of actuaries analyzes questions like these to support recommendations to insurance companies to help them address different data scenarios that will allow them to better understand the risk of their policies, as well as meet customers there. where they are and where the data is.

Finally, what impact can new hires have at Arity?

Chandrawanshi: It’s not just the tools and technologies we use that make it exciting for software engineers. We encourage people to pursue what interests them. We encourage rotation within teams so they can broaden their horizons and work on solving new problems, learning new technologies and creating new products.

chao: As a data scientist or analyst, there is no better place than Arity. There are a lot of interesting problems to solve. There really is big data to look at. There’s a stack of modern technology you can use and learn from and a great group of people to work with. You immediately see the impact on our business and, more importantly, on our customers and users.

Source link


Comments are closed.