Research
Lindell S. Cumes
Co-founder · Durable Intelligence
My research focuses on durable AI: how intelligent systems can preserve commitments, learn from experience and carry useful progress across long-running work.
I study the architectures and evaluations needed for AI agents to continue working reliably as context, tools and models change. Nenya is an initial application of this research to software teams.
Research interests
AI agents, long-horizon reasoning, continual learning, agent memory, AI reliability, evaluation and neurosymbolic systems.
Education
- Massachusetts Institute of Technology
Bachelor of Science in Computer Science and Engineering
Electrical Engineering and Computer Science - Imperial College London
MSc Applied Machine Learning
Electrical and Electronic Engineering