OdorPi raises £700K pre-seed to give AI a sense of smell
What's the deal? London-based deeptech startup OdorPiDealroom has a profile for this one. Try Dealroom → has raised £700,000 in pre-seed funding to develop technology that detects disease-related compounds through smell. ZincDealroom has a profile for this one. Try Dealroom → and SFC CapitalDealroom has a profile for this one. Try Dealroom → co-led the round, with additional backing from Innovate UKDealroom has a profile for this one. Try Dealroom →.
What's the endgame? Founded in 2025 by Koye Sodipo, the company is building an electronic nose that pairs sensor hardware with AI models to detect and classify volatile organic compounds (VOCs) — the molecules behind smell. Its initial focus is non-invasive healthcare screening.
How it works: The system analyses VOCs from urine, aiming to spot signs of disease before symptoms appear. Longer-term targets include urinary tract infections and certain cancers.
Central to the approach is an in-house sampling platform designed to collect and label up to 5,000 smells per day, feeding training data to the models. In early testing using the Sniffin' Sticks olfactory protocol, OdorPi says its models correctly classified scent categories 84.15% of the time across 12 categories.
What they say: "Olfaction provides a non-invasive, passive and continuous medium to screen for a number of diseases," said Sodipo. "Our mission is to enable clinicians and individuals catch disease even before symptoms appear."
Where the money goes: OdorPi will use the funding to develop its sensor chemistry, expand smell data collection and labelling, and continue validating its diagnostics platform with clinical partners. Its scientific team includes researchers with PhDs in sensor chemistry and electronics, backed by clinical advisors.
The signal: At £700,000, this is a modest pre-seed even by early-stage standards — its amount sits in roughly the bottom half of comparable rounds. But it reflects growing investor appetite for deeptech that translates hard science into diagnostics, betting that a sense clinicians have long relied on informally can be measured, digitised, and scaled.
Read more: Tech.eu
Image credit: Generated with Gemini
