Tiliter, a startup originating in Australia, has secured $7.5 million in Series A funding, spearheaded by Investec Emerging Companies. The company specializes in computer vision technology designed to enable cashierless checkout systems, effectively eliminating the necessity for product barcodes.

Established in 2017, Tiliter leverages artificial intelligence for recognizing retail products, offering benefits like the ability to automatically identify loose items – such as fresh fruits and baked goods – without manual staff intervention. The system also contributes to a reduction in packaging waste.

The AI-driven product recognition system also minimizes errors in product selection, whether those errors are intentional or accidental in nature.

According to co-founder and COO Martin Karafilis, “A significant challenge arises from items lacking barcodes, leading to slower and less efficient manual identification processes.” He explains this applies to bulk goods, produce, bakery items, and mixed selections, as well as situations where barcodes are obscured or damaged.

“Furthermore, a substantial amount of plastic is produced globally for barcodes and identification packaging. Our technology has the potential to significantly decrease, and even eliminate, single-use plastic for retailers.”

Currently, the company’s primary focus is on the supermarket sector, where its product identification system boasts over 99% accuracy in less than a second.

Tiliter has developed specialized hardware that can be easily integrated with existing checkout systems to operate the computer vision technology, providing retailers with a readily deployable cashierless solution.

The company’s website highlights that its AI software utilizes its own proprietary data and does not collect in-store information. It is capable of functioning with bagged items and can distinguish between even subtle sub-categories, such as different varieties of tomatoes or apples. The system can also identify organic produce based on retailer-specific indicators.

“Our software is pre-trained,” Karafilis states, addressing the question of system adaptation to a retailer’s inventory. “We’ve concentrated on developing a versatile and scalable software solution that functions seamlessly for all retailers immediately. If an item is not currently recognized, a supermarket can add it in approximately 20 minutes, and the system possesses self-learning capabilities.”

Beyond its ease of installation, facilitated by the hardware’s compatibility with existing retail infrastructure, Tiliter asserts that its solution is more cost-effective than other autonomous store options currently available. (Amazon is a prominent competitor in this space.)

The hardware is sold directly to retailers, accompanied by a yearly subscription fee for the software, which includes a commitment to 24/7 global service and support.

“We supply dedicated hardware – including a camera and processor – that can be easily retrofitted to any existing checkout, scale, or point of sale system, integrating our vision software with the point of sale,” Karafilis explains, noting that the pandemic is increasing demand for easily implemented cashierless technologies.

The startup points to a 300% increase in the adoption of ‘scan and go’ technology in the US over the past year, attributed to COVID-19, and anticipates continued global expansion.

While specific customer numbers are not being disclosed, early adopters of Tiliter’s AI-powered product recognition system include Woolworths in Australia, with over 20 stores currently utilizing the technology; Countdown in New Zealand; and several retail chains in the US, such as Westside Market in New York City.

The Series A funding will be used to accelerate expansion into Europe and the US, with “numerous” supermarkets expected to implement the technology in the coming months.