We build AI and privacy technologies, developed in house. TERNIA is our current focus.
Nebula Technology Lab builds AI and privacy technologies. Our current lead product is TERNIA, which helps pharma teams prioritise high-confidence molecule candidates from smaller assay datasets, reducing reliance on costly screening campaigns and heavy data infrastructure.
Before AI is useful in drug discovery, someone has to pay for the data. Most models need millions of historical data points before they can say anything useful about a compound. TERNIA works the other way round. It takes a single small assay and returns a ranked shortlist, with a confidence flag on every call, so teams act on the strong ones and set the rest aside.
The wider Nebula stack includes ICPP, our privacy layer, which we are extending to address a problem pharma teams know well: assay data is valuable enough that companies are careful about sharing it. The Nebula Portfolio System, our work in explainable portfolio construction, is at an earlier stage. Both are developed in house.
Developed in-house, these technologies are at different stages of research and development, each backed by academic white papers and preprints, some already publicly available.
TERNIA finds the winners before the expensive screening starts. It is built on COGENT3, our AI engine architecture, in which many small agents form temporary teams around three-agent building blocks, test approaches, keep what works and adapt over time. Useful patterns are remembered and reused, so the system can take on new edge cases without reworking the whole model.
See more at ternia.bioEvery result starts with the data behind it. We work from the assay data a pharma or biotech team already holds, cleaning and structuring it so the signal that matters is easy to find, rather than assuming more data is always the answer. Because our systems learn from small datasets, teams reach high-confidence candidates without the cost and delay of a large screening campaign or heavy data infrastructure.
Portfolio construction and analysis software, built on quantitative models. The interface is designed to take an objective stated in plain English and return a modelled allocation with the reasoning behind it, bounded by the constraints it was given. The design aim is consistency, explainability and bounded risk in how an allocation is arrived at.
Organisations sitting on valuable data face a common problem: taking part in a shared system usually means revealing more than they want to. ICPP is a privacy protocol built to separate participation from exposure, so a party can take part without disclosing what it holds. We are now extending it to assay and dataset confidentiality in drug discovery.
Nebula Technology Lab Limited is registered in the Dubai International Financial Centre and was founded in 2025.
Nebula Technology Lab LimitedThese organisations have supported the company and its founders.
ECOSYSTEM
Nebula Technology Lab is a member of the Innovation Hub.
INCUBATOR
Nebula Technology Lab is part of the LvlUp Labs founder programme.
NETWORK
Nebula Technology Lab is supported by Ignyte.
For enquiries about Nebula Technology Lab or any of the technologies on this page, use the form or write to contact@nebulalab.ae.