Silicon Software Lead

About the role

Normal Computing is building a new class of ASIC that computes using stochastic analog dynamics in memory. The Silicon Software Lead will head the team that creates the software stack – compiler, runtime, kernels, drivers and hardware abstraction layer – that turns the chip into a programmable platform for real‑world AI inference workloads.

Key responsibilities

• Lead and grow a multidisciplinary silicon software team covering compiler, runtime and systems software.
• Define and own the architecture of the stack from ML framework ingestion through compilation, scheduling, memory management and execution on Normal hardware.
• Co‑design with silicon architects to shape ISA definition and hardware abstraction so the chip is fully programmable.
• Drive development of runtime, kernels, drivers, profiling and debugging tools for inference workloads.
• Maintain continuity from simulation and FPGA prototypes to silicon, ensuring software never stalls hardware tape‑out.
• Set the silicon software roadmap, align milestones with the hardware program and recruit the engineers needed.

Required profile

• Proven experience building software stacks for accelerators or non‑standard hardware (compilers, runtimes, kernels, drivers).
• Track record of leading engineers while staying hands‑on in design and code.
• Strong systems programming background with deep understanding of ML inference workloads.
• Comfort working on hardware that evolves from simulation through bring‑up.

Required skills

• C++
• Rust
• Python
• MLIR
• LLVM