Postdoctoral Researcher: Hardware Acceleration Edge AI

The CAI-3 group within the Computing and Artificial Intelligence Division at Los Alamos National Laboratory (LANL) is seeking a Postdoctoral Research Associate to advance hardware architectures that support trustworthy, resource-efficient AI at the edge. You will design, implement, and evaluate FPGA/HLS-based hardware accelerators for computations that support real-time confidence estimation, fault and anomaly detection, and robust decision-making in autonomous systems operating under strict latency, power, and reliability constraints.

You will explore hardware-software co-design methodologies that bring statistical and machine learning reasoning closer to the sensing and computing hardware. Your work will include benchmarking prototype hardware against existing software and hardware baselines for latency, resource utilization, and numerical accuracy. You will contribute to closed-loop evaluation on a multimodal autonomy testbed under degraded or GPS-denied sensing conditions. The successful candidate will join an interdisciplinary team of scientists and engineers developing creative solutions to enable reliable autonomous decision-making for national security applications  will provide opportunities to support your growth as a research scientist, including conference travel, manuscript publication, networking, service, and mentorship opportunities.

CAI-3 (Information Sciences) develops methods and tools for understanding complex interactions and extracting actionable information from massive data streams, from novel deep learning architectures to algorithms that exploit revolutionary hardware such as quantum and neuromorphic circuitry. This position sits at the intersection of two CAI-3 teams:
Novel Computing, which develops custom coprocessors (ASICs, FPGAs, VHDL) to accelerate critical computations for LANL's mission, and Mission Integrative Computer Science, which bridges FPGA design, signal processing, and hardware-software co-design to deliver custom hardware systems for NNSA Defense Programs and Global Security applications. You will work alongside researchers in AI/ML, uncertainty quantification, and computer engineering on hardware approaches that support confidence-aware AI reasoning.

You will work alongside researchers in AI/ML, uncertainty quantification, and computer engineering on hardware approaches that support confidence-aware AI reasoning.

• PhD in Electrical Engineering, Computer Engineering, Computer Science, or closely related field, completed within the last 5 years (or requirements completed and degree pending)
• Hands-on experience with FPGA design flows, RTL (Verilog/VHDL) and/or high-level synthesis (Vitis HLS, Vivado, or similar)
• Experience with digital hardware/accelerator architecture design and implementation
• Proficiency in C/C++ and at least one hardware description or HLS language
• Experience integrating custom hardware accelerators with a software or embedded processing stack
• Strong written and verbal communication skills, and demonstrated ability to work in a multidisciplinary research team

Education/Experience
: A Ph.D. in Electrical and Computer Engineering, Computer Science, Artificial Intelligence, Mathematics, Materials Science, Physics, or related physical sciences, completed within the last 5 years or to be completed soon.

• Experience with algorithm-hardware co-design, particularly mapping numerical or statistical computations onto FPGA or custom accelerators
• Familiarity with reduced-precision/quantized numerics and their effect on algorithm accuracy
• Background in machine learning, probabilistic methods, or uncertainty quantification
• Experience with robotics/autonomy pipelines involving SLAM, visual-inertial odometry, or sensor fusion
• Experience designing for bounded-latency or real-time system constraints (safety-critical, avionics, or similar domains)
• Track record of benchmarking custom hardware against CPU/GPU baselines
• Demonstrated experience in conducting original scientific research through peer-reviewed publication records. Publication record in relevant venues (FPGA, FCCM, DAC/ICCAD, robotics, or ML-hardware conferences)
• Prior experience at a national lab or in safety-critical systems development
• Ability to obtain a Q Clearance, which requires U.S. Citizenship

Outstanding candidates may be considered for a Director's Fellowship and for the prestigious Darleane Hoffman, Richard P. Feynman, or J. Robert Oppenheimer…