Senior Software Engineer

You will design and implement high-performance LLM inference software in modern C++. You will optimize vectorization, memory access, threading, scheduling, and data movement across x86 CPUs and FPGAs. You will profile bottlenecks, improve transformer execution, integrate quantization methods, and maintain comprehensive automated tests.

Responsibilities

• Design and implement high-performance LLM inference software
• Develop and optimize C++ libraries
• Optimize SIMD threading and memory hierarchy utilization
• Coordinate data movement and offloading between CPUs and FPGAs
• Optimize model execution through vectorization cache efficiency and scheduling
• Develop performance profiling tools and methodologies
• Apply NUMA-aware memory management
• Implement token streaming KV-cache optimization and batching
• Integrate quantization sparsity and mixed-precision techniques
• Add unit performance acceptance and regression tests

Requirements

• 7+ years of C++ software development experience
• Knowledge of C++ templates and memory management
• SIMD programming and intrinsics-based vectorization experience
• High-performance computing numerical computing or ML inference optimization experience
• Knowledge of tensor computation and memory access optimization
• Knowledge of multithreading NUMA and CPU optimization
• Systems software development profiling and benchmarking experience
• Experience with hardware accelerators and software-hardware interfaces

Benefits

• Company-paid medical dental and vision insurance for employees and dependents
• Company-paid life and disability coverage
• Supplemental hospital critical illness and accident coverage
• Unlimited paid time off
• 13 paid company holidays
• Remote-first work
• Company-provided computer and home office setup
• Equity
• 401(k) matching from day one