Silicon Architect

About the Company

Normal Computing builds silicon that turns thermal noise from an obstacle into a computational resource. We co-design the full stack: AI-native EDA systems and advanced ASICs that provide 10-100× more AI inference per dollar and per watt.

Responsibilities

• Define the architecture and microarchitecture of novel AI accelerator compute blocks, including PE array design and datapath organization
• Translate workload analysis and research findings into hardware specifications and microarchitecture documents
• Reason across the full stack to defend PPA tradeoffs involving memory hierarchy, interconnect, and physical design
• Partner with compiler leads on ISA co-design to ensure the compute tile is programmable and compilable
• Direct block-level pre-silicon validation using FPGA and cycle-accurate simulation
• Stay current with AI accelerator research to articulate the advantages of Normal's approach

Requirements

• Degree in Electrical Engineering, Computer Engineering, Computer Science, or equivalent experience
• Substantial experience in architecture or microarchitecture of high-performance digital systems like AI accelerators
• Ability to translate algorithm-level workload analysis into hardware specifications
• Experience with simulation-driven architecture using cycle-accurate or analytical models
• Familiarity with quantization and reduced-precision approaches for inference
• Experience writing microarchitecture specifications for RTL implementation
• Proficiency in Python or C++ for performance modeling
• Familiarity with SystemVerilog or equivalent RTL