The challenge
Angad Sandhu, working at his startup-venture studio Stylite, was running a structured cross-channel search for an early-stage AI engine to commercialize. Their criteria was specific: high TRL, validated by proof-of-concept, addressing a known industry pain point, and capable of being applied across multiple sectors.
This kind of search usually runs into a familiar set of frictions. TTO portfolios live across hundreds of separate institution websites with no consistent format. Cold outreach to academic groups can take months to get a reply, if it lands at all. Conferences concentrate activity into expensive high-jeopardy windows. And once a candidate technology does surface, making contact with the team and keeping the conversation moving is a serious undertaking for a small outfit.
The solution
Angad joined Inpart's partnering ecosystem and started reviewing matches surfaced through the platform in response to their technical interests. Among them was MFeaST, a technology being developed by researchers at Queen's University.
MFeaST helps organizations evaluate complex datasets, including both directly related data and seemingly unrelated data sources. It surfaces the variables, patterns, and hidden relationships that matter most for prediction, risk assessment, optimization, and decision support. Corexa believes that its ability to identify meaningful relationships across seemingly unrelated variables differentiates MFeaST from traditional complex-dataset intelligence platforms.
At its core, MFeaST leverages a heterogeneous ensemble architecture where multiple predictive engines evaluate the same problem from different perspectives. A proprietary decision layer then reconciles these outputs into a unified prediction. The technology was originally validated in oncology, where it identified 17 microRNA biomarkers that distinguished 15 types of neuroendocrine neoplasms with 98% accuracy.
"Inpart's platform covered our discovery thesis. It allowed us to review multiple technologies in a short period before we found the one we were most interested in"
Once Angad found a technology to pursue, Inpart’s platform engagement team made the introduction to the TTO at Queen's University. After initial online introductions, an early in-person meeting then brought Angad's team together with the Queen's TTO officers and inventors.
"It helped to introduce us to one another in a way that a virtual meeting might not have accomplished to this degree"
From there, Inpart stayed involved through the months of conversations that followed, checking that requirements were being met and chasing delays on either side. When the NDA was signed, the Inpart team congratulated the collaborators on reaching their first major milestone.
"This communication momentum was extremely important, especially as we moved toward negotiating the licensing process. The Inpart platform team helped us stay organized, maintain communication, and keep track of the innovation and technology pipeline"
The result
exclusive technology license
In early 2026, Corexa Technologies Inc. was incorporated as the exclusive licensee of MFeaST.
Corexa is currently refining use cases across fintech, medtech, and AI for ESG / green initiatives.
In fintech, MFeaST is being evaluated for credit-risk and market-intelligence applications. By assessing financial indicators alongside relevant external signals - such as trade conditions, macroeconomic shifts, supply-chain relationships, and other contextual factors - it can help teams focus on the variables most relevant to default risk and market trends. Corexa is also exploring how MFeaST can complement proprietary financial-market prediction capabilities, with further details remaining confidential & in-development.
In AI for ESG / industrial technology, MFeaST is being applied to complex operational datasets, including BOS reactor studies in steelmaking. The technology can identify the variables associated with high-CO2 operating conditions and oxygen consumption, while supporting a broader assessment of operational and contextual factors beyond conventional sensor data. This work is intended to inform process optimization, emissions reduction, and sustainability outcomes.
MFeaST was originally used in peer-reviewed journals for finding biomarkers for NEN cancers. Corexa is developing a proof of concept that explores demographic clustering of disease susceptibility and the potential for earlier, more targeted prevention. The company is also assessing opportunities in rare diseases affecting pets and animals.
Across these areas, Corexa is focused on identifying the strongest use cases and building evidence for broader commercial deployment. The company welcomes conversations with industry partners, data owners, and research groups interested in exploring well-defined applications of MFeaST.
"Our timeline is to find the strongest use case, then negotiate a multi-year extension of the license, followed by raising capital"