Positron AI 4 дня назад
Performance & Debug Software Engineer – AI Inference
200 000 – 300 000 $в год
- Удалённо
PythonC#FirmwarePerformance OptimizationBenchmarkingTritonProfilingDistributed InferenceLLM InferenceLinux perfPerfettoVFIO
Описание
Positron AI is looking for a Performance and Debug Software Engineer to profile and optimize end-to-end LLM inference on custom AI accelerators and build benchmarks and regression tests for distributed inference and KV cache.
- LLM inference
- Linux perf
- Distributed Inference
- VFIO
- Profiling
- benchmarking
- Performance optimization
- C#
- Triton
- firmware
- Python
- Perfetto
- FPGA
- debugging
- eBPF
- Linux
- HIP
- ASIC
- CUDA
- KV cache