Postdoctoral Research Associate - Neuromorphic Computing

As a postdoctoral researcher, you will help solve some of the most challenging problems the world faces. You will perform impactful research on a range of significant problems, and you will apply your work in multidisciplinary domains alongside globally recognized experts. You will bring creative thinking, teamwork, and hardware-software co-design skills to bear as you develop new methods to address scientific and engineering problems, collaborate with leaders in your field and across the laboratory, and disseminate innovative results through publications and presentations.

The Learning Systems Group at Oak Ridge National Laboratory advances the next generation of artificial intelligence by co-designing algorithms and computing systems to address complex challenges in science, energy, national security, and health. Our researchers develop scalable machine-learning models, deep-learning architectures, and data-driven methods for platforms ranging from edge sensors and neuromorphic systems to national supercomputers and quantum computers. Our work includes scientific knowledge discovery, autonomous and energy-efficient systems, threat intelligence and data fusion, agent-based modeling and digital twins, and predictive analysis for environmental and health applications.

We are committed to recruiting and retaining highly motivated, creative researchers, postdoctoral fellows, and students who will help develop intelligent systems that learn, reason, and operate effectively across diverse computing environments.

• Research, develop, and optimize novel low-level software, drivers, and firmware for executing SNNs on resource-constrained embedded platforms.
• Research, develop, and evaluate hardware-software co-design approaches for edge neuromorphic systems and event-driven processing architectures.
• Implement and validate spiking neural network (SNN) execution frameworks on microcontrollers (ARM, RISC-V), FPGAs, or specialized neuromorphic processors.
• Conduct power profiling, energy-per-inference measurements, and latency benchmarking on physical edge hardware platforms.
• Build strong collaborations within ORNL and with the neuromorphic computing and embedded systems communities worldwide.
• Publish research in high-impact journals and conferences.
• Author peer-reviewed papers, technical papers, reports, and proposals for internal and external release.
• Represent ORNL and the Learning Systems Group by giving talks at public forums (conferences, workshops, guest lectures, etc.).

• Ph.D. in Computer Engineering, Electrical Engineering, Computer Science, or a closely related field.
• Strong programming proficiency in C, C++, and Python.
• Hands-on experience with embedded systems programming, bare-metal development, or Real-Time Operating Systems (RTOS such as FreeRTOS or Zephyr).
• Experience with hardware development tools and microcontrollers based on ARM or RISC-V architectures, and with FPGA development using Hardware Description Languages (Verilog, VHDL, or System Verilog).
• Experience with standard hardware communication protocols (e.g., SPI, I2C, UART, PCIe).
• Experience in software engineering, including writing efficient, scalable software, using version control tools such as Git, maintaining software, writing software documentation, and creating tutorials explaining how to use the software.
• Solid understanding of neuromorphic computing and SNNs, including common neuron models (e.g., leaky integrate-and-fire) and SNN training or conversion methods (e.g., spike-timing-dependent plasticity (STDP), surrogate-gradient learning or ANN-to-SNN conversion).
• A strong publication record in peer-reviewed journals and conferences demonstrating sound embedded hardware/software or computational research.
• Excellent written and oral communication skills.
• Ability to work independently and to participate creatively in collaborative teams across the laboratory.
• Ability to set priorities to accomplish multiple tasks within deadlines and adapt to changing needs.

• Experience with physical power profiling tools, oscilloscopes, and logic analyzers for hardware energy benchmarking.
• Experience interfacing neuromorphic hardware or SNNs with event-based vision sensors (DVS), edge sensors, or robotics platforms.
• Familiarity with event-driven hardware interfaces and protocols, such as Address-Event Representation (AER).
• Experience with hardware-aware AI optimization, including quantization, pruning,…