Compiler Backend Engineer
Normal Computing · Mountain View, CA · New York, NY
Infrastructureonsite$200k–$300k
At a glance
Highlights
- Build custom compiler backend mapping LLM inference onto in-memory hardware.
- Own scheduling, fusion and memory planning measured by end-to-end latency.
- Validate optimizations on simulators and FPGA prototypes, then on silicon.
What they offer
- Equity
Ideal candidate
You are a performance-driven compiler engineer who loves working close to novel hardware and designing optimizations from first principles to make LLM inference fast.
Requirements
- Deep experience with compiler optimization or code generation for accelerators or non-CPU targets
- Comfort working close to hardware reasoning about memory bandwidth and utilization
- Willingness to design compilation approaches from first principles for novel hardware
- Track record of shipping optimizations that improved measured performance
- Comfort building software for evolving pre-silicon hardware from simulation to bring-up
- Fluency in Python and familiarity with C, C++, or Rust
Lab profile
About Normal Computing
Applied AI company building foundational hardware and software for semiconductors and AI infrastructure.
What they build
- Foundational hardware for semiconductors
- Software for semiconductor industry
- Critical AI infrastructure
Known for
- Applied AI for silicon engineering
- ASIC architecture and design verification
- FPGA prototyping and validation
Key facts
- Operates across New York City, Mountain View, and London
Posted · Last seen