A deep tech startup named Apoha, focused on developing AI models to engineer new types of substances, has exited its stealth phase, announcing a $36 million Series A funding round. This significant investment is earmarked to advance its unique approach to understanding and manipulating materials, ranging from pharmaceutical compounds to food products and industrial paints. The London and San Francisco-based company’s Series A was spearheaded by European venture capital firm Singular, with additional participation from Draper Associates, and continued support from existing seed investors including Redalpine, Seedcamp, Wilbe, and Nucleus. Further bolstering its development, Apoha has also received a grant from Innovate UK, the United Kingdom’s national innovation agency.
At the core of Apoha’s technology is an innovative method for gathering data on how materials behave, a departure from traditional AI models often trained on text or image data alone. The company posits that the key to unlocking the creation of novel materials lies in analyzing the waveforms generated when substances are suspended in liquid and subjected to external forces. These unique wave patterns, Apoha has found, correlate directly with a material’s properties, encompassing characteristics like taste, smell, and reactivity. This “liquid intelligence” approach, as Apoha terms it, aims to provide its AI models with the capacity to suggest precise modifications or entirely new material compositions to achieve desired attributes.
Anshika Srivastava, Apoha’s co-founder and Chief Operating Officer, articulated this distinctive vision, noting that while machines have mastered visual and textual comprehension of matter, they have yet to grasp its sensory and kinetic qualities. The company seeks to bridge this gap, enabling AI to “taste, smell, or feel matter,” thereby perceiving how a drug dissolves, a flavor endures, or a material degrades. Shamit Shrivastava, a mechanical engineer with post-doctoral research from the University of Oxford and a PhD from Boston University, co-founded Apoha in 2021 alongside Anshika, a former Goldman Sachs banker. Shamit, who serves as Apoha’s CEO, holds the patent for the liquid waveform analysis that underpins the company’s data generation, along with several specialized hardware devices developed for their experiments.
The company’s name, derived from a Sanskrit word meaning “negation or exclusion,” reflects a philosophical stance that objects are often best defined by what they are not. This principle subtly informs their technological approach. Apoha has engineered a laboratory device capable of analyzing a minuscule material sample, suspending it in liquid, and then applying controlled physical stresses. The device records the resulting wave patterns, yielding over 1,000 distinct numerical descriptors of the material’s behavior in a matter of minutes, a process that conventionally takes days or even weeks. This readout, dubbed VIBE (Variations in Inter-facial Behaviour Under Excitation), constitutes Apoha’s initial commercial offering. These raw recordings are then translated into what Shamit Shrivastava describes as a “behavioral embedding,” a numerical fingerprint that AI models can learn from and compare.
The predictive power of VIBE measurements is considerable, according to Apoha’s founders. It can forecast a drug’s stability within the body, assess whether a plant-based protein will mimic the texture of chicken meat, or predict the wear characteristics of a new material over time. One early success story involved assisting a food company in rapidly sourcing a replacement for a key ingredient in its plant-based vegan chicken product after a supplier unexpectedly ceased operations. In the pharmaceutical sector, the technology holds immediate promise for screening drug candidates before they enter costly clinical trials. A multi-year research partnership with German pharmaceutical firm Boehringer Ingelheim, for instance, demonstrated Apoha’s ability to identify high-risk antibody candidates with over 90% accuracy using minimal material.
Beyond pharmaceuticals, Apoha is collaborating with German biotech Ethris to predict the behavior of lipid nanoparticles carrying mRNA, similar to those used in certain COVID-19 vaccines, within animal systems. The company also maintains partnerships with Somru BioSciences and several Fortune 500 companies across the pharmaceutical, food and beverage, and materials industries, having completed approximately 40 customer projects to date. With a team of around 25 employees, Apoha intends to deploy the Series A funds to scale its platform, encompassing both its custom hardware for VIBE data acquisition and the AI models built upon this data, to accommodate a broader range of sample types and a growing customer base. Raffi Kamber, co-founder and general partner at Singular, commented that Apoha exemplifies a new generation of European scientific companies where AI is not merely a future aspiration but a practical tool actively transforming biological research.
