ASIC Architect for AI Accelerator — Efficient LLM Inference

Normal Computing Corporation; New York (NY); Híbrido; USD 200.000 - 320.000; Jornada completa

Descripción de la vacante

Normal Computing is seeking a Hardware ASIC Architect to define silicon and system microarchitecture for a novel AI accelerator platform, targeting large energy efficiency gains for LLM inference. This role leads hardware/software co-design with collaboration across compiler, RTL, and analog teams.

You will translate transformer architectures and diffusion execution flows into custom tiles, memory hierarchies, and interconnects, delivering throughput-per-watt on real workloads.

Formación

• Degree in Electrical Engineering, Computer Engineering, Computer Science, or equivalent.
• Experience in architecture of high-performance digital systems: AI accelerators or compute engines.
• Fluency moving between algorithm-level analysis and hardware specification.
• Experience with simulation-driven architecture and cycle-accurate models.
• Familiarity with quantization and reduced-precision inference.
• Experience writing microarchitecture specifications and working with RTL engineers.
• Proficiency in Python or C++ for performance modeling, and familiarity with SystemVerilog RTL.
• Comfort working in exploratory architecture where the answer is not yet known.

Responsabilidades

• Define silicon and system microarchitecture for AI accelerator compute blocks.
• Translate workloads into hardware specifications and produce microarchitecture documents for RTL.
• Defend full-stack PPA tradeoffs across algorithm, memory, interconnect, and physical design.
• Co-design ISA with the compiler lead and ensure alignment.
• Direct pre-silicon prototyping with FPGA teams and de-risk tapeout decisions.
• Stay current with AI accelerator research and articulate differences of Normal's approach.

Conocimientos

Architecture design

C++/Python

SystemVerilog RTL

Simulation-driven design

Quantization

Co-design

Performance modeling

Educación

Electrical Engineering or related

PhD welcome but not required

Herramientas

RTL engineering

Descripción del empleo

Normal Computing is seeking a Hardware ASIC Architect to define silicon and system microarchitecture for a novel AI accelerator platform, targeting large energy efficiency gains for LLM inference. This role leads hardware/software co-design with collaboration across compiler, RTL, and analog teams.

You will translate transformer architectures and diffusion execution flows into custom tiles, memory hierarchies, and interconnects, delivering throughput-per-watt on real workloads.