AIONICS enables rapid materials discovery and cell performance prediction

AIONICS partners with leading battery and electronics manufacturers to bring the power of machine learning to battery research and development.

What is AIONICS?

Motivated by the importance of high performance energy storage to meeting our energy challenges, and frustration with the existing pace of innovation in battery technology, we have built AIONICS: a machine learning (ML)-based platform for accelerating the design and deployment of batteries. AIONICS (pronounced "a-i-onics", for "A.I." + "ionics") leverages large data sets and advanced ML techniques to enable companies to move beyond the “black box” view of batteries as complex and unpredictable systems. The AIONICS platform has two main thrusts: the first for assisting in battery material design for battery manufacturers, and the second for optimizing battery usage in devices for manufacturers of battery-powered devices. By providing a better understanding of how battery performance depends on the relevant input variables, AIONICS enables the accelerated generation of battery-related intellectual property for the former and improves the reliability and longevity of devices for the latter. The mission of AIONICS is to provide a simple and cost-effective platform for companies to take advantage of the tremendous power of machine learning and data-driven techniques, without needing to hire a team of engineers to do so. We want to democratize ML-driven battery innovation.


Materials Design for battery manufacturers

Promising battery materials must simultaneously exhibit many properties, and the routes to optimizing these properties are often not clear. Optimizing materials for these properties requires disentangling complex structure-property relationships that may not adhere to preexisting scientific intuition. The result is that optimization often entails “turning the knobs” at one’s disposal (i.e. material composition and synthesis procedures) in a guess-and-check manner until the solution is found. This is a slow, manual process that can significantly hamper the rate of valuable intellectual property generation. ML offers a valuable approach to these problems by learning directly from measured data, circumventing the need for complete scientific understanding of the underlying phenomena. By combining your proprietary data with publicly available data from the scientific literature, AIONICS’ materials optimization toolkit makes it easy to leverage the power of big data and machine learning.



The second focus area of AIONICS is in optimizing battery use for device integration. In this case, the optimization is based on maximizing battery lifetime. Cycling identical batteries in different ways yields a large variance in cycle life that is often not intuitive to understand. However, each unique battery type (determined by the specific composition and even the particular manufacturer) exhibits telling signatures of these events in their cycling data that can provide substantial predictive power. By aggregating commercial cell cycling data from our partners across the industry, AIONICS’ ML algorithms learn to predict capacity fade in commercial cells directly from the time series cycle data on cell voltage and capacity. This enables quick and easy battery control optimization around the specific power demands of your device. This helps device manufacturers spend less time and money searching for optimal battery control algorithms and lets them get back to what they want to do: building their device.


AIONICS is a Rho AI company powered by years of experience in the machine learning and energy spaces. 


Dr. Austin Sendek, CEO

Ph.D. in Applied Physics, Stanford

Screen Shot 2019-01-16 at 1.27.54 PM.png

Dr. Lenson Pellouchoud, Engineering Contractor

Ph.D. in Materials Science, Stanford


Michael Chau, Engineer

Undergraduate, UC Berkeley


Throop Wilder, Advisor

Co-Founder, 24M Technologies, Crossbeam Systems, American Internet Corp.; Williams College B.A.


Joel Moxley, Advisor

Co-Founder, Foro Energy, Biota Technology, Rho AI; Adjunct Professor, Stanford University; MIT Ph.D.

stuart-macmillan (1).png

Stuart Macmillan, Advisor

Former Chief Scientist, National Renewable Energy Laboratory; Adjunct Professor, Stanford University; Stanford Ph.D.



We work in close partnership with our customers' technical teams to help them build better batteries and devices. Most commonly, we leverage the AIONICS platform as an internal tool to deliver bespoke, highly customized insights to our customers. These engagements provide the industry knowledge, finances, and test cases that are essential to building a valuable platform. 

Every company is different, and so are every one of our partnerships. Our customers range from technology giants and small startups. Working under strict confidentiality, we work on your timeline, with your existing infrastructure, to get you the insights you need. None of our engagements are ever the same - every project requires a unique set of data and seeks answers to unique questions. If we don't have the data you need, we'll get it; if we don't have existing tools to answer your questions, we'll build them. As AIONICS continues to develop into a highly flexible and user-friendly platform, we will deploy on-site with customers.

Interested in learning more or working together? We want to hear from you!