Deep-tech startup Apoha has emerged from stealth with $36 million in Series A funding, betting that the next breakthrough in AI for medicine, food and materials wont come from bigger language models but from a new data class entirely: liquid wave-form measurements of how matter behaves.
Inside The Round
The round was led by European deep-tech firm Singular, with participation from Draper Associates and existing seed backers Redalpine, Seedcamp, Wilbe and Nucleus. London- and San Francisco-based Apoha was founded by Oxford physicist Shamit Shrivastava and former Goldman Sachs trader Anshika Srivastava; the company has been operating quietly since its 2021 incorporation.
What Liquid Intelligence Is
Most AI models for biology and chemistry today work from sequence data (e.g., genome, protein) and structure data (e.g., crystallography, cryo-EM). Apohas thesis is that there is a third, missing modality: measurements of the wave forms substances generate when suspended in a liquid and acted on by outside forces. The company calls the resulting data layer Liquid State Intelligence and the AI built on top liquid intelligence.
Where It Applies
Apoha is targeting AI-designed materials across pharma and biotech (proteins, drug formulations), food and beverage and industrial materials like paints. The platform is designed to give partners a way to design and predict how a new substance will behave not just what it should look like before any of it is synthesised at scale.
Part Of A New AI Frontier
The launch arrives in a week where deep-tech AI rounds dominate the boards alongside Coralogixs $200M observability Series F and morphs soft-robotics stealth exit in London. With CVPR 2026 opening in Denver, Apohas pitch is a clean break from frontier-model maximalism: that some of the biggest AI gains of 2026 will come from building entirely new data classes, not training larger LLMs on existing ones.
Reporting based on coverage from Fortune, TechFundingNews and Apohas company release.