Objective

Aqueduct was founded with a clear mission: to make it easier for businesses to connect with university researchers and access the expertise they need for innovation.

Developed in partnership with the University of Strathclyde, the platform uses AI and conversational search to match industry enquiries with relevant researchers. By analysing institutional research databases like PURE, Aqueduct transforms the complex process of academic matchmaking into a fast, intuitive experience.

Before working with LOOPSIO, the matchmaking process relied on manual questionnaires and email exchanges that took weeks. Aqueduct set out to prove that with AI and automation, these interactions could happen in minutes.

‍

‍

What we delivered

• Developed an AI-powered chatbot interface that enables users to submit enquiries and instantly receive matches to relevant university researchers.
• Built a vector database that uses research abstracts, bios, and keywords to generate semantic matches with natural-language queries.
• Created an embeddable JavaScript widget that integrates directly into the University of Strathclyde’s “Work With Us” portal.
• Ensured security and scalability, implementing authentication tokens, CORS validation, and responsive design for a seamless user experience.
• Delivered a live demonstration of the platform’s capabilities, enabling Aqueduct to showcase its value to university stakeholders and potential investors.

‍

Our Approach

LOOPSIO worked closely with the founder, Ross Methven, to translate a research-driven concept into a scalable digital product. The development process combined AI experimentation with a practical product mindset, focusing first on building a working proof of concept and then transforming it into a deployable widget for live testing.

The collaboration followed two phases. The first introduced large language models (LLMs) and vector databases for semantic search, marking LOOPSIO’s first project involving AI integration. The second phase refined the system into an embeddable and production-ready component, ready for use on institutional websites.

Throughout both phases, the team maintained close communication with Strathclyde University’s technical and communications teams to ensure compatibility, compliance, and alignment with institutional infrastructure.

‍

Challenges We Overcame

• First AI Integration: Implementing LLMs and vector databases for semantic matching required new technical research and internal learning.
• Complex Data Sources: Integrating with Strathclyde’s PURE system and handling sensitive institutional data demanded precision and security.
• Widget Deployment: Converting the chatbot into a responsive, lightweight embeddable script was technically challenging but achieved successfully.‍
• Balancing Innovation with Stability: Delivering a functional live demo required balancing experimentation with robust engineering.

‍

Results

• Functional Proof of Concept: Delivered a working AI chatbot that demonstrates Aqueduct’s ability to match enquiries with relevant researchers.
• Successful Integration: The chatbot widget is now embedded and tested on the University of Strathclyde’s site, showcasing real-world impact.
• Scalable Foundation: The codebase and architecture are ready for expansion to additional universities across the UK and internationally.‍
• AI Capability Established: LOOPSIO successfully integrated LLMs into production, marking a milestone in its technical evolution.

‍

Conclusion

Aqueduct is a new way of discovering expertise, built on the belief that collaboration should be simple, intelligent, and accessible.Through close collaboration and technical innovation, LOOPSIO helped transform a complex manual process into a scalable, AI-powered platform.

The project not only delivered a tangible proof of concept for the University of Strathclyde but also positioned Aqueduct for future growth, paving the way for universities worldwide to bridge the gap between research and real-world innovation.

‍