Behind the Idea: Tensorflight | The Fintech Times

Despite the adoption of digital technology, insurance remains one of the most traditional sectors in the financial industry. Continued use of legacy systems has hampered their development. But organizations like Tensorflight are doing more now to ensure the industry catches up.

Tensorflight delivers structured real estate data to the insurance industry by instantly analyzing aerial, satellite and street view imagery. Robert Kozikowski Co-Founder and CPO of the company. He has over 10 years of experience in technology and is now leveraging his expertise to accelerate the digitization of insurance.

Robert Kozikowski, co-founder and CPO of Tensorflight

We interviewed Mr. Kozikowski to learn more about the company.

Tell us more about your company and its products

Tensorflight is an AI imaging-based insurtech that automates commercial property inspection and claims processing. Our mission is to make high quality data available to the insurance industry to enable informed, timely and accurate decision making.

Automate the time-consuming and costly process of in-person property inspections using three image sources: satellite, aerial, and ground level. We also assist insurers in evaluating commercial and residential properties and risks as part of the underwriting process. Finally, it helps you better understand your portfolio exposure.

We are a Polish/US insurtech with offices in Warsaw, New York and London. We are proud to work with many of the world’s largest commercial property insurance companies, including: Zurich Insurance Group When QBEs.

What problem was your company founded to solve?

Face-to-face property inspections are time consuming and costly. The commercial real estate industry is currently underserved and has to make decisions based on inaccurate real estate infrastructure data. These are the problems we are solving.

As previously mentioned, we use data from three imagery sources – satellite, aerial and ground level, as well as public, contributed and proprietary sources – to provide detailed coverage of 99% of properties in developed countries. We can provide.

Our solution can also automatically evaluate various property attributes such as building footprint, structure type, roof pitch and shape, number of floors, and detect the presence of ACP panels.・Fire spread to Grenfell Tower.

Our AI and computer vision technology allows us to look at the same image and determine what a human assessor could do. Our solution enables clients to significantly reduce the cost and time of inspections in the insurance process. On average, the cost of in-person property inspections can be reduced by more than a quarter, and our solutions can greatly improve the efficiency of your billing function.

How has your company evolved since its launch?

Our company was founded in 2016 with a very scientific and engineering mindset. We are PhD holders and ex-Google Engineers looking to apply our expertise to deep learning. The first breakthrough came in his 2019. nephila capital, the first company in the insurance industry to invest in our product. We continued to invest in research and engineering, gradually not only developing our products, but building the company we are today.

Series A funding QBE Ventures Last October was a big milestone for us. Since then, we have signed many strategic partnerships that have allowed us to further strengthen our offerings.For example, this year Airbus Defense and Space Intelligence Businessis a global leader in geospatial data, intelligence and defense solutions. This has allowed us to strengthen our ability to analyze and value real estate.

This partnership gives us access to the highest commercial resolution imagery available on the market. It also benefits from data with the highest geolocation accuracy. State-of-the-art satellite imagery allows us to collect building-related information and detect property changes.

This ranges from information on roof condition and asbestos detections, to damage analysis and assessment after natural disasters such as hurricanes, to inside conflict zones where on-site inspection is not possible and other imaging sources cannot provide a level of detail. increase. Required for insurance procedures.

We have also recently appointed and hired several notable senior positions. Jakub Dryasis one of the longest-serving members and a former head of growth responsible for spearheading the company’s expansion across key markets in the US, UK, Europe and Australia, was promoted to CEO over the summer. We look forward to the next chapter of our business under his leadership.

What was your biggest challenge or most “hard moment” to overcome?

Digital transformation in the insurance industry is one of the biggest challenges. It is still in its infancy and legacy systems continue to hinder progress. The pandemic has underscored the need for property and casualty insurance sectors to adopt technological innovations, change how property inspections and insurance processes are managed, and reduce friction in the customer experience.

As a result, insurers are increasingly looking for faster, streamlined alternatives and are beginning to adopt new technologies such as AI and machine learning that provide richer data insights and faster, more efficient service. You can see the change in thinking.

What is your greatest accomplishment or “proudest moment” so far?

We are proud of our pro bono work. Kyiv Economic School (KSE) Institute Provision and analysis of satellite imagery to assess damage caused to infrastructure within Bucha, Irpin and Mariupol.

Since the start of the Ukrainian War in February 2022, the KSE Institute, a leading think tank and analytical arm of the Kyiv University of Economics, has Initiative Supported by the Ukrainian government. This initiative is an open-source data project to analyze information about damage inflicted on civilian infrastructure, effectively assessing and tracking the cost of war.

Analyzing satellite imagery and rooftop geocoding information has provided insight into the extent of damage to specific buildings within the analyzed area, while also providing the exact geographic coordinates of the building. Our team of AI experts trained a deep neural network model to detect and classify war-damaged buildings in satellite imagery. To date, approximately 70,000 buildings in Mariupol, Irpin and Bucha have been identified and analyzed using our model.

The project is also helping the Ukrainian government assess the country’s recovery requirements. For example, intelligence can help determine whether to prioritize rebuilding power plants for the winter months, or rebuild hospitals and schools based on the needs of residents in disaster areas.

The next plan is to analyze satellite images covering Lyschansk, Melitopol and Popasna. However, satellite imagery is very costly. More than 5 million euros are needed to scan her 100,000 square kilometers of Ukraine awaiting damage assessment. Additionally, this data becomes outdated quickly as the damage spreads, so we are currently looking for sponsors and partners to help fund and sustain the project.

How would you describe your company culture?

Tensorflight has an enterprising culture and we are a meritocracy that values ​​creativity, ideas and innovation. We believe this approach allows us to stay on the cutting edge and uncover the best ideas. It also keeps staff motivated and engaged while developing leadership.

As a company, we are informal and have a “people first” approach. It also has a flat structure where everyone on the team has the right to voice their opinion. They are encouraged to share them openly and often. Also from the insurance sector, the focus is on reliability, security, consistency and professionalism.

What is in store for the future?

we have big plans. We invest more not only in product development, but also in developing our people, hiring the best talent, and expanding our global reach. In terms of product development, we are currently working on improving the accuracy of several key attributes we evaluate, such as geocoding and area.

Our plan is to enable digital insurance for commercial underwriting and make residential digital insurance more efficient and cost effective. We work to enable all the essential pieces an insurance company needs.

We are also preparing to raise a Series B funding next year. This allows us to further expand and enhance our offerings.

Leave a Reply

Your email address will not be published. Required fields are marked *