Developing antibody drugs begins with finding an antibody that can attach to a specific spot, known as an epitope, on a protein linked to a disease. Finding an antibody that fits that spot can be difficult.
Bengaluru-based immunitoAI is using artificial intelligence to tackle this problem. Its platform, AbInitio, designs antibodies for a chosen target and epitope, which the startup then tests in its own lab.
Founded in 2020 by Dr Aridni Shah and Trisha Chatterjee, the startup began when the two met in August that year through Entrepreneur First, a talent accelerator where individuals look for potential co-founders.
Shah, the CEO, holds a PhD in biological sciences from NCBS-TIFR and a bachelor’s degree in biomedical sciences from the University of Delhi. Chatterjee, the CTO, has a master’s degree in computer science from the University of Delhi and previously worked as a senior software development engineer in R&D at GreyOrange.
The idea grew out of Shah’s PhD research on honeybees. Since honeybees are not a model organism, obtaining antibodies against their proteins was difficult. Early discussions with industry experts revealed that drugmakers faced a similar challenge in finding suitable starting antibodies, Chatterjee says.
Chatterjee spent about three months developing the idea and up to three weeks building the first prototype. That version ranked potential antibodies, a simpler task than the antibody design work AbInitio handles today.
Designing and testing antibodies
ImmunitoAI’s biology team first selects a disease area, a target protein, and the epitope the antibody needs to bind to. It also determines whether the antibody should block or activate the target. AbInitio then designs antibodies and antibody-like structures, including nanobodies, which are small, single-domain antibodies, for the chosen epitope.
The platform trains on 3D structures of binding protein pairs from the Protein Data Bank, a public repository. A mathematical transformation developed by the startup converts the atoms’ 3D coordinates into two graphs, with atoms represented as points and links showing where they interact.
The model learns how the shapes of two proteins fit together, with the grooves of one accommodating the bumps of another. It also learns which amino acids, the building blocks of proteins, bind to each other. The platform checks drug-like properties during the design stage to reduce the need for rework later.
“We take roughly 1,000 to 10,000 designs to the lab, depending on the strategy. An AI platform can generate binders, but which of them really bind in the real world is what matters,” Chatterjee says.
The designs are tested in the lab to identify which antibodies bind to their targets. Further experiments assess their drug properties.
The startup is currently pre-revenue, and its drug candidates remain at the preclinical stage. However, Shah says candidates tested in cell-based assays have shown functional relevance and potency.
Its programmes include CD40, GITR and CXCR4, covering disease areas such as oncology, autoimmune disorders, haematology, and neurology.
Licensing as a business model
ImmunitoAI plans to work with pharmaceutical companies and generate revenue in two ways: by out-licensing its own drug candidates or by using its platform to identify a final candidate for a partner's chosen target.
Partners can stop the process at any stage or select from candidates the startup has already developed.
The company is also developing its own candidates for targets relevant to the pharmaceutical industry. Chatterjee says the work generates data that can be used to improve the platform.
The startup’s latest funding was a Series A round of Rs 52.14 crore led by Ashish Kacholia, with participation from 3one4 Capital. Its total funding stands at Rs 66.98 crore, and its investors include pi Ventures.
ImmunitoAI has plans to expand into the US and Europe. Over the next year, it will upgrade AbInitio and advance its own drug candidates for out-licensing.
“We hope the AI platform improves many times over, and we might need to test very few candidates after this enhancement,” Chatterjee says.
ImmunitoAI is part of YourStory's Tech30 cohort—a showcase of India's most promising startups of 2026, unveiled at TechSparks Bengaluru.
Edited by Teja Lele