Y Combinator-backed Holy Grail is using machine learning to build better batteries

For a long, long time, renewable energy proponents have considered advancements in battery technology to be the Holy Grail of the industry.

Advancements in energy storage has been among the hardest to achieve economically, thanks to the incredibly tricky chemistry that’s involved in storing power.

Now, one company that’s launching from Y Combinator believes it has found the key to making batteries better. The company is called Holy Grail and it’s launching in the accelerator’s latest cohort.

With an executive team that initially included Nuno Pereira, David Pervan and Martin Hansen, Holy Grail is trying to bring the techniques of the fabless semiconductor industry to the world of batteries.

The company’s founders believe that the only way to improve battery functionality is to take a systems approach to understanding how different anodes and cathodes will work together. It sounds simple, but Pereira says the computational power hadn’t existed to take into account all of the variables that go along with introducing a new chemical to the battery mix.

“You can’t fix a battery with just a component,” Pereira says. “All of the batteries that were created and failed in the past. They create an anode, but they don’t have a chemical that works with the cathode or the electrolyte.”

For Pereira, the creation of Holy Grail is the latest step on a long road of experimentation with mechanical and chemical engineering. “As a kid I was more interested in mechanical engineering and building stuff,” he says. But as he began tinkering with cars and became fascinated with mobility, he realized that batteries were the innovation that gave the world its charge.

In 2017 Pereira founded a company called 10Xbattery, which was making high-density lithium batteries. That company, launching with what Pereira saw as a better chemistry, encapsulated the industry’s problem at large — the lack of a holistic approach to development.

So, with the help of a now-departed co-founder, Pereira founded Holy Grail. “He essentially told me, ‘Do you want to take a step back and see if there’s a better way to do this?’ ” said Pereira.

The company pitches itself as science fiction coming from the future, but it relies on a combination of what are now fairly standard (at least in the research community) tools. Holy Grail’s pitch is that it can automate much of the research and development process to create new batteries that are optimized to the specifications of end customers.

“It’s hard for a human to do the experiments that you need and to analyze multidimensional data,” says Pereira. “There are some companies that only do the machine-learning part and the computational science part and sell the results to companies. The problem is that there’s a disconnection between experimental reality and the simulations.”

Using computer modeling, chemical engineering and automated manufacturing, Holy Grail pitches a system that can get real test batteries into the hands of end customers in the mobility, electronics and utility industries orders of magnitude more quickly than traditional research and development shops.

Currently the system that Holy Grail has built out can make 700 batteries per day. The company intends to  build a pilot plant that will make batteries for electronics and drones. For automotive and energy companies, Holy Grail says it will partner with existing battery manufacturers that can support the kind of high-throughput manufacturing big orders will require.

Think of it like bringing the fabless chip design technologies and business models to the battery industry, says Pereira.

Holy Grail already has $14 million in letters of intent with potential customers, according to Pereira, and is expecting to close additional financing as it exits Y Combinator.

To date the company has been backed by the London-based early-stage investment firm Deep Science Ventures, where Pereira worked as an entrepreneur in residence.

Ultimately, the company sees its technology being applied far beyond batteries as a new platform for materials science discoveries broadly. For now, though, the focus is on batteries.

“For the low volume we sell direct,” says Pereira. “While on high-volume production, we will implement a pilot line through the system… we are able to do the research engineering with the small ones and test the big ones. In our case when we have a cell that works, it’s not something that works in a lab, it’s something that works in the final cell.”