Sucafina – a leading sustainable Farm to Roaster, Swiss-based coffee company – and ProfilePrint – a Singapore-based AI-driven predictive and prescriptive food ingredient profiler – have announced a co-investment in Csmart, a Brazil-based developer of AI-powered image recognition technologies for green coffee grading.
“We feel confident that Csmart’s proven image recognition technology will bring shared value to producers, importers, and QC professionals alike. This is not about replacing people but empowering QC professionals to become even more productive,” says Dave Behrends, Managing Partner & Head of Trading at Sucafina.
Founded by producer, quality control (QC) professional, and software developer Francisco Massucci Silveira, Csmart designs AI-powered image recognition technologies for coffee grading that aim to improve the coffee industry for producers, professionals, and roasters alike. Csmart’s latest image recognition green-grading machine-learning technology can rapidly scan a sample of green coffee beans and accurately identify defect count within minutes. This technology could help QC professionals by reducing repetitive tasks, improving and accelerating communication between QC teams, and reducing the carbon emissions from shipping countless green coffee samples to buyers and sellers across the world.
“Ultimately, we hope that collecting and analyzing data on a large scale will unveil new and valuable information about coffee production, enabling the incorporation of data-driven decisions throughout the coffee value chain,” Massucci says.
Currently, Csmart’s software can identify the total defect count and confirm screen size of a sample of green coffee. The forthcoming iteration of this software will be able to CCc quantify each individual defect by type and take image scans of each identified bean. One of the main benefits of this technology is that it cuts down on the time and resource inefficiencies caused by sending green samples back and forth across the globe. With Csmart’s technology, green coffee can be scanned by the seller and an objective report delivered to the buyer for their review. This can significantly speed up the time it takes to approve a sample and reduce the carbon footprint of coffee sales.
This new image recognition green grading technology will be complementary to ProfilePrint, which incorporates data from multiple sources to enable users to rapidly ascertain the suitability of ingredients without traditional sensory evaluation.
Alan Lai, founder and CEO of ProfilePrint, adds: “Csmart’s image recognition technology provides additional key data points for our global ProfilePrint solution, bringing us closer to our goal of a device-agnostic data platform. It also increases the range and accuracy of our prediction capability, empowering buyers and sellers across the supply chain to increase productivity and significantly reduce carbon footprint.”
Csmart offers a variety of benefits to actors across the supply chain, including producers. By making objective green grading more available, Csmart will help democratize information about coffee quality.
“Producers can better understand the product that they have and the faster approval time can mean that they get paid more quickly,” says Tim Heinze, Coffee Education Manager at The Center, an online global learning community focused on creating a more equitable and innovative industry.
Sucafina is a leading sustainable Farm to Roaster coffee company, with a family tradition in commodities that stretches back to 1905. Today, with more than 1,200 employees in 32 countries, the Group is one of the leading coffee trading houses in the world and one of the few focused entirely on coffee. Sucafina’s extensive supply chains span from producer-facing export operations to destination sales offices around the world.
Csmart is an AgTech startup focused on specialty coffees with the intention of automating processes and creating business intelligence through the implementation of computer vision, artificial intelligence, and data analysis. Ultimately, we hope that collecting and analyzing data on a large scale will unveil new and valuable information about coffee production, enabling the incorporation of data-driven decisions throughout the coffee value chain.
ProfilePrint is an AI-driven predictive and prescriptive food profiler. Its patented food fingerprinting technology provides users with accessible, affordable and portable solutions to predict outcomes at the point of use, within seconds. ProfilePrint’s solution allows sellers and buyers to ascertain the agreed quality of food ingredients digitally, expediting existing assessment processes and facilitating trade – from the farm all the way to end-buyers. ProfilePrint is a Singapore-based company with a global footprint.