Spore.Bio raises $23M to apply machine learning to microbiology testing

Recalls in the food and beverage industry due to contamination incidents can have catastrophic effects. Not only do companies have to pay fines and damages, but the impacts on the brand’s reputation can be long-lasting.

That’s why Spore.Bio, a Paris-based deeptech startup, is trying to reinvent microbiology testing to avoid the next PR crisis in the food industry. After raising an €8 million pre-seed round ($8.3 million at current exchange rates) a little bit more than a year ago, the company just secured a $23 million Series A round.

Singular is leading the round. Point 72 Ventures, 1st Kind Ventures (the family office of the Peugeot family), Station F and Lord David Prior are also participating. Existing investors LocalGlobe, No Label Ventures and Famille C are putting more money in the company as well.

The reason why Spore.Bio managed to raise so quickly after its pre-seed round is that there’s real customer interest. The startup has already signed a few commercial contracts that can cover up to 200 factories. Spore.Bio had to open a waitlist to make sure it can keep up with demand.

So what makes Spore.Bio’s technology special? In the food and beverage industry, microbiological tests require several days. Companies have to take a sample and send it to a specialized lab for testing.

“Picture this, we’re in 2022, everything is hyper-optimized. You’ve got lean manufacturing everywhere, every step is optimized and counted in minutes to get a result, to move from one step to the next,” co-founder and CEO Amine Raji told TechCrunch. “And bam, you’ve got a 5-day imponderable test in the agri-food sector, and 14-day test in the pharmaceutical and cosmetics sectors, to get a result because you have to wait for the bacteria to grow.”

First, testing has to happen offsite because petri-dish-based testing involves demultiplying any potential bacteria. So you can’t risk contaminating other parts of the factory with your testing. Second, the bacteria incubation part takes time.

Spore.Bio is using a completely different process. The company sends light at specific wavelengths toward a sample and records the spectral signature. Thanks to a pre-trained deep learning algorithm, it can detect whether that specific sample contains any bacteria or pathogens.

That model is Spore.Bio’s most important asset. The startup has signed a partnership with the Pasteur Institute to access its biobank of bacteria samples.

In the coming months, it wants to manufacture testing machines that customers can use directly in their own factories. As a result, microbiology testing can happen directly on site. The company claims it reduces the overall process from days to a matter of minutes.

Before founding Spore.Bio, Raji was a food and beverage manufacturing engineer working for Nestlé. He naturally focused on the industry he already knew. But it turns out that microbiology testing is much larger than anticipated.

Companies manufacturing cosmetic products have also expressed interest in Spore.Bio’s technology. “Manufacturers need to get rid of preservatives due to customer demands, environmental concerns and other reasons. Except that preservatives are bacteria-killing preservatives,” Raji said.

Similarly, the pharma industry found a use case for its most advanced treatments. “There is a growing need, especially for innovative therapies, such as gene and cell therapy,” Raji said. He added that these products tend to have a short shelf life, which can be as low as seven days. So these therapies can’t go through the usual testing processes in such a short timeframe.

With today’s funding round, the startup expects to significantly grow its team. There are currently 30 people working for the company, and they will be 50 by the end of 2025.