Staff Research Scientist, AI-Hardware Co-Design

Responsibilities

• Lead the co-design of AI algorithms and hardware architecture to define optimal compute strategies for intelligent systems.
• Drive technical execution from initial architectural design through to validated proof-of-concept.
• Perform feasibility analysis using high-fidelity simulation and modeling to quantify the impact of memory hierarchy, dataflow, and precision on performance.
• Optimize neural architectures to exploit specific hardware capabilities, balancing power and area constraints across various compute substrates.
• Collaborate with business leads and domain experts to identify high-impact opportunities for integrated sensing and actuation systems.
• Provide thought leadership by staying current with the evolving AI hardware landscape and mentoring junior engineers.

Requirements

• PhD in Computer Architecture or Integrated Circuit Design specialized in AI workloads.
• 3+ years of industry experience applying architectural principles to real-world engineering constraints.
• Deep expertise in hardware-aware deep learning and model optimization (quantization, tiling, and dataflow).
• Proficiency with modern machine learning frameworks such as PyTorch or JAX.
• Strong understanding of memory hierarchy, on-chip networks, and execution models.

Preferred Qualifications

• Proven experience with silicon execution, such as taping out a complex SoC or custom AI accelerator.
• Fluency in Digital Signal Processing (DSP) and its intersection with deep learning.
• Familiarity with hardware construction languages like Verilog, SystemVerilog, Chisel, or PyMTL.
• A strong publication record in top-tier academic conferences or journals.

About the Company

Analog Devices (ADI) is a global semiconductor leader that bridges the physical and digital worlds to enable breakthroughs at the Intelligent Edge. Operating through the Analog Garage—our innovation lab in downtown Boston—we combine analog, digital, AI, and software technologies to solve high-impact problems in automation, robotics, healthcare, and energy.