We are hiring for a research assistant or postdoc to do energy efficient AI, spiking neural networks and neuromorphic computing, with me, Christos Bouganis and George Constantinides at Imperial College London. Deadline for applications is May 15th. Feel free to email me informally, or apply at the link below. https://www.imperial.ac.uk/jobs/search-jobs/description/index.php?jobId=27677&jobTitle=Research+Assistant+/+Research+Associate+in+Energy-Efficient+AI,+Spiking+Neural+Networks+and+Neuromorphic+Computing Some more details below (also at the link above). Dan Goodman About the role We are seeking a Research Assistant (a Masters graduate wanting to undertake a PhD) or Research Associate (a PhD graduate wanting to undertake a postdoc). Our aim is to explore new approaches to energy-efficient artificial intelligence based on temporal neural computation. The project investigates how spiking neural networks, temporal coding, and learned neural delays can enable accurate computation with extremely low-precision weights, potentially unlocking a new generation of ultra-efficient AI hardware. Working at the intersection of machine learning, computational neuroscience, and digital hardware design, the successful candidate will contribute to developing neural architectures in which time (spike timing and delays) replaces numerical precision and memory movement as the key computational resource. The project will combine algorithm development with hardware-aware modelling and evaluation on FPGA-based platforms. The successful candidate will have the opportunity to help define new approaches to temporal neural computation and energy-efficient AI. This role provides an exciting opportunity to work on fundamental questions in neuromorphic and energy-efficient AI, while developing techniques that could translate into future AI accelerators and edge-AI technologies. The role will be affiliated with NeuroWare, the new national Innovation and Knowledge Centre in Neuromorphic Computation, providing the post-holder access to a rich network of industrial and academic collaborators and routes to direct impact. Pre-doctoral candidates are strongly encouraged to apply. Candidates appointed as Research Assistant will have the opportunity to register for a PhD during the appointment, subject to standard university procedures. What you would be doing In this role, you will investigate neural architectures that exploit temporal coding and learned delays to enable efficient computation on digital hardware. You will develop and evaluate spiking neural network models, explore training methods for delay-based computation, and analyse how temporal representations trade off with weight precision and memory usage. You will primarily be collaborating with three Imperial College academics: Prof Christos Bouganis, Prof George A. Constantinides and Dr Dan Goodman, who each bring leading expertise relevant to this research programme. You will implement models in software frameworks for neural simulation and machine learning, and work with hardware-aware efficiency metrics to evaluate energy, memory, and latency trade-offs. The project will also involve exploring how these architectures map to digital hardware platforms, including FPGA-based systems.