Description
Normal Computing is hiring a Hardware ASIC Architect to define the silicon and system microarchitecture for a custom unconventional compute platform designed for LLM and diffusion model inference. The role leads hardware/software co-design, translates transformer and diffusion workloads into compute tiles, memory hierarchies, and interconnects, co-designs an instruction set, performs full-stack performance, power, and area tradeoffs, and directs block-level pre-silicon validation. Candidates should have substantial experience in high-performance digital system architecture or microarchitecture, familiarity with quantization and reduced-precision inference, and proficiency in Python or C++ for performance modeling and SystemVerilog or equivalent RTL.
