Positron
Software Engineer, Senior
$150K–$250K
Remote (United States) · Remote · Senior · Full-time · Posted Feb 11, 2025
- Experience
- 7+ years of professional C++ software development; open to hiring at a more senior level for the right candidate.
- Visa sponsorship
- Candidates must currently be authorized to work in the U.S.; no new visa sponsorship is available, but H-1B transfers may be facilitated for eligible candidates.
About the job
Develop high-performance C++ software for running and optimizing open-source LLM inference on custom FPGA and x86 CPU hardware.
What you'll do
- Design and optimize inference software, libraries, and model execution for custom hardware.
- Work with hardware and systems engineers on data movement, computational offloading, profiling, and performance optimization.
- Implement system-level LLM optimizations and integrate quantization, sparsity, and mixed-precision techniques; maintain testing in CI.
What they ask for
- Strong C++ templates and modern memory-management skills, including professional C++ development focused on performance-critical applications.
- Hands-on SIMD programming and intrinsics-based vectorization experience.
- Experience in HPC, numerical computing, or ML inference optimization, including model execution and tensor or memory-access optimization.
- Knowledge of multithreading, NUMA, CPU optimization, systems profiling and benchmarking, and hardware accelerators.
Nice to have
- Familiarity with LLVM/Clang or GCC compiler optimizations.
- Experience with LLM quantization, sparsity, and mixed precision; distributed inference, networking, graph partitioning, or execution scheduling.
Pay
$150,000 – $250,000 USD base salary
Tools
C++17/20, SIMD, AVX-512, SSE, FPGAs, x86 CPUs, perfetto, VTune
AI focus
Optimize LLM inference on custom hardware, including token streaming, KV cache, batching, quantization, sparsity, and mixed-precision execution.
Model-written summary · the original posting is authoritative