My personal highlight of NVIDIA’s GTC Europe Conference in Munich was yesterday’s Inception Awards, where five of Europe’s hottest deep learning and AI startups pitched in front of an expert jury and in front of NVIDIA founder and CEO Jensen Huang. The finalists were selected from among the 700 European startups participating in NVIDIA’s Inception program, which accelerates the development of startups involved in AI and deep learning.
In the end, it was Gamaya, a 20-person startup from Switzerland, which won the pitch competition and took home the prize – the AI supercomputer DGX Station. Founded in 2015, Gamaya helps farms improve their productivity and sustainability, enabled by hyperspectral imaging and artificial intelligence. In order to learn more about the promising startup, we sat down with Luca Baldassarre, Gamaya’s Head of Data Science, who pitched at yesterday’s event and won the supercomputer for his company.
Luca, please give us a short introduction what Gamaya does and how it all started.
Gamaya helps agricultural businesses to significantly improve the efficiency and sustainability of their operations and production. We combine a number of sophisticated technologies including drone-based hyperspectral imagery, weather stations, satellite data and ground measurements. Our solutions include early detection of plant diseases and weeds, optimisation of soil treatment and fertilisation, as well as prediction of yields.
Gamaya is a spin-off of EPFL, the Swiss Polytechnic in Lausanne, after years of environmental research using ultralight airplanes and hyperspectral imaging.
Who are your biggest competitors and what differentiates Gamaya from their solution?
We have a number of competitors that do field-level analysis, such as Airinov in France and Slantrange in US, as well as large-scale analytics including Descartes Labs and Cibo Technologies. We are different in that we seamlessly bridge small-scale with large-scale imaging combining drone imaging with satellite data, developing crop-specific analytics.
Furthermore, our analytics are developed with a much more sophisticated hyperspectral camera that we designed, developed and calibrated in-house.
I’m sure you had to invest a substantial sum in research and development. How much venture capital did you raise so far and who are the investors?
We raised CHF 3.2 millions in 2016 and are preparing for Series B in 2018. The investors are: the Sandoz Family Foundation, the Peter Brabeck Family Office, Seed4Equity and VI Partners.
You just won a supercomputer at the NVIDIA Inception Awards. What are you going to do with it?
Play Doom! Just joking. 🙂 We are currently training several deep learning models for detecting agronomical issues in sugarcane and tobacco. The DGX Station will dramatically shrink our development time and boost the models’ complexity and accuracy.
With Gamaya you’re addressing a global market. Which countries/markets are you going to focus on in the coming year and what are the next milestones you’re planning to achieve?
In the next year, our commercial focus will be on sugarcane and tobacco in Brazil. We are also conducting a number of R&D projects to develop solutions for major crops in Eastern Europe. These highly industrialized crops provide a great opportunity to bring immediate value by solving several efficiency and productivity bottlenecks in their cultivation.
Besides Gamaya, yesterday’s Inception Awards also showcased the following four AI startups: Automotive Artificial Intelligence, a virtual 3D platform that replicates a realistic city environment for testing automotive driving software; Valossa, an AI software that understands video like a human; Restb.ai, an AI startup helping the real estate industry identify and classify images; and Deckard.ai, a startup which brings machine learning and process mining to project revolutionize management.