Engineering Living Factories with AI

Cells are nature's most powerful factories, capable of producing almost any molecule imaginable. From common household goods like soap and cosmetics, to life-saving medicines, we rely on microorganisms like yeast and bacteria for countless products. The persistent challenge, however, lies in coaxing these cells to produce the things we need at commercial scale. Artificial intelligence could hold the key to using the intricacy of biology to transform the way we make the world’s products.

A-Life is X’s moonshot to unlock the manufacturing power of biology. We’re tackling the immense complexity of engineering cells to be our living factories, with the goal to make things cheaper, faster, and with less waste than traditional manufacturing.

The Search for Biological Systems Intelligence

My path to A-Life started in a surprising place: as a product manager for Google Play search and discovery. I worked on systems that relied on the collaboration of hundreds of engineers developing interrelated algorithms to deliver the right information instantly and reliably to millions of users. I was inspired by the way Google builds products based on cutting-edge algorithms that stay cutting-edge year after year.

Before joining Google, I earned my PhD in Computational Biology. I worked in the lab, and I worked on algorithms for biology. In biology, changes cause a cascade of chain reactions. You can't change one thing without causing ripple effects everywhere else. This is a systems problem. Yet life sciences are so complex that teams typically treat each layer separately, changing only one variable at a time (like temperature or a single gene) to see what it does.  It’s a slow, manual process that can take years. 

I came to X to bring together these two worlds: computational systems thinking and the dynamic, layered complexity of biology. Our team – a small group of synthetic biologists, systems engineers and AI experts – set out to accelerate the manufacturing potential of biology. We knew that meant giving scientists better tools to get the most out of their real-world scientific datasets, and help them optimize across millions of potential experiments. So we built an AI platform designed specifically for biology.

Biomanufacturing in the Age of AI

Biomanufacturing is the process of using cells like yeast and bacteria to produce materials on an industrial scale. For example, the pharmaceutical, wellness, and beauty sectors rely on biomanufacturing to make vitamins, cosmetics and medicines.

This is not a new idea: Humans have been harnessing cells to make things for us for millennia, beginning with ancient fermentation practices to make bread, beer and wine. If you’ve ever made bread, you know that sourdough starter is a living community of wild yeast and bacteria, which you feed with flour and water to turn into dough. You place the dough in a warm part of your kitchen, and the yeast does its work to make the bread rise. 

In biomanufacturing, scientists follow a similar process: they select cells to produce one specific thing, like insulin.  These cells are fed a carefully prepared diet, usually a simple sugar, which makes the cells grow and produce more of the desired product. Instead of a  warm kitchen, scientists place the cells in a bioreactor, where they can control variables like temperature, nutrient levels, pH and oxygen. The challenge is to find the perfect balance between all these different variables to get the most out of the cells.