DesMind Labs
DesMind Labs
Our research, experimental projects and collaborations.
Focus areas
Research areas
Privacy-preserving AI
Models that learn from sensitive data without exposing it. This is the core of our PhD research.
Federated learning
Training across hospitals, sites and devices while raw data stays where it was created.
Responsible AI
Evaluation, governance and ethics designed into systems, not added after launch.
Conversational systems
Assistants that keep context, cite sources and hand over to a person when they should.
Large language models
Retrieval, fine-tuning, structured outputs and guardrails for dependable behaviour.
Digital health
AI for early detection and support in care, starting with perinatal mental health.
Live now
Live demonstrations
Research assistants, in-browser machine learning and interactive walkthroughs.
Open the demo hub- DesbotCited RAG assistant over an indexed knowledge base
- Des Lab BotOpen-domain LLM assistant with file analysis
- Desbot BriefRule-based, knowledge-driven assistant
- Computer vision and NLP toolkitTensorFlow.js classification, sentiment and summarisation
- Federated learning walkthroughInteractive FedAvg, privacy and fusion simulations
Research
Publications
Peer-reviewed research in AI for healthcare, power electronics and the life sciences.
Read
Scoping review of governance, privacy, and design in AI-supported conversational agents for perinatal mental health
Preprint · OSFA five-level nested neutral point-clamped inverter topology
IET Power ElectronicsA single-phase, nine-level switched-capacitor-based inverter
Journal of Power ElectronicsStator winding chording in five-phase synchronous reluctance motors
Journal of Energy Systems
Open source
Open source
MeetUp: developer connection platform
A social platform that connects developers with shared interests.
View codeNext.js AI chatbot
A full-stack AI chat application built with Next.js and TypeScript.
View codeGovernance review analysis code
Reproducible notebooks and data behind the conversational-agent governance review.
View codeResearch collaboration
We welcome research collaborations, pilot deployments and joint projects, especially in digital health and privacy-preserving AI.