Nondeterministic polynomial-time hard (NP-hard) optimization problems can be mapped on a novel class of Ising processors, typically implemented in CMOS exploiting digital compute or mixed signal compute-in-memory techniques. However, current implementations do not allow to scale up to larger problem sizes due to limited graph connectivity, inter/intra memory data movement cost, and processing element and random data generation overhead. This results in an exponential time-to-solution and energy consumption when scaling up problem size.In this project, we aim to overcome these bottlenecks by smartly integrating compute-in-memory architectures exploiting emerging p-bit devices, which are compact and able to generate random numbers at a high pace. In this project, the p-bit enabled compute-in-memory fabric will be integrated into a complete digital processor, including hardware software co-design necessary to come to a complete p-bit-enabled Ising processor for combinatorial optimization problems.We are seeking a highly motivated postdoctoral researcher to address these issues and help build the next generation of processors capable of solving combinatorial optimization problems.The ideal candidate has expertise in processor architectures, digital design, and AI acceleration. This role will be mentored by a KU Leuven professor, Prof. Marian Verhelst, in association with the imec Compute System Architecture department, as part of a dynamic collaboration between imec and the university.As the Successful CandidateYou will offer research and development support to advanced research on novel architectures for AI compute accelerators exploiting emerging devices.Specifically, your primary responsibility will be:Architecture specifications for a p-bit enabled Ising accelerator, defining the hardware intrinsics and interfaces towards integration into an overall compute system.Architecture optimization and digital circuit design.Analysis of the standalone accelerator in terms of the performance and power consumption for the critical kernel(s) in the Ising compute applications.Exploration of accelerator integration into a multi-core Ising system.Collaborating with multi-disciplinary teams e.g., device, circuit, and algorithm teams, is pivotal. You will closely engage with different experts to design and develop the compute-in-memory-based architectures. You will lead research activities on designing architecture specifications for combinatorial optimization networks, algorithm benchmarking, system architecture definition, and end-to-end architecture development and evaluation. Regular participation and contributions to projects and team meetings are expected.What we do for youWe offer you the opportunity to join one of the world’s premier research centers in nanotechnology at its headquarters in Leuven, Belgium. With your talent, passion, and expertise, you’ll become part of a team that makes the impossible possible. Together, we shape the technology that will determine the society of tomorrow.We are committed to being an inclusive employer and proud of our open, multicultural, and informal working environment with ample possibilities to take initiative and show responsibility. We commit to supporting and guiding you in this process; not only with words but also with tangible actions. Through imec.academy, 'our corporate university', we actively invest in your development to further your technical and personal growth.We are aware that your valuable contribution makes imec a top player in its field. Your energy and commitment are therefore appreciated by means of a market-appropriate salary with many fringe benefits.Who you areA Ph.D. in electrical engineering, AI/ML, applied physics, with relevant experience.Proven research experience in processor architectures, hardware-software co-design, AI/ML accelerators.Experience or familiarity with AI algorithms, digital architecture design, compute-in-memory architecture, system architecture modeling and simulation, and data analysis tools.Strong communication, interpersonal, and organizational skills to bridge research groups and project partners in an interdisciplinary team.Excellent verbal and written communication skills, fostering seamless interaction with local and remote colleagues and partners.You have the capability and mindset to flexibly adapt your research to new application challenges and are eager to learn about new topics. You have a strong vision to propose novel solutions to such challenges as well as the technical skills to develop and implement them.Adaptable to interdisciplinary team collaboration.This opportunity invites you to be part of pioneering research and innovation within an international, diverse, and dynamic environment.This postdoctoral position is funded by imec through KU Leuven. Because of the specific financing statute which targets international mobility for postdocs, only candidates who did not stay or work/study in Belgium for more than 24 months in the past 3 years can be considered for the position (short stays such as holiday, participation in conferences, etc. are not taken into account).
#J-18808-Ljbffr