Senior System Software Engineer – Accelerated Microservices and Libraries
Core
Design and develop accelerated microservices and libraries for real-time vision-language model inference, serving video ingestion, preprocessing, and model serving for harnesses and agents.
Role type
Senior System Software Engineer (AI Inference)
Builds
Real-time vision-language model inference microservices and optimized libraries
Domain
Computer Vision / Video AI / GPU Inference
Deliverable
production ML models
Required skills
C/C++, Python, Microservices architecture, Computer Vision, Vision-Language Models, GPU Inference, Real-time video processing, Performance optimization, Debugging
Preferred skills
NVIDIA DeepStream, TensorRT, Triton Inference Server, CUDA, RTSP, FFmpeg, GPU profiling (Nsight), x86/ARM deployment, KV-cache management, Quantization
Technologies
C, C++, Python, CUDA, TensorRT, Triton, DeepStream, FFmpeg, RTSP, Nsight Systems
Responsibilities
Design and develop accelerated microservices and libraries; Integrate and serve computer vision and vision-language models in production; Profile and optimize end-to-end latency, throughput, and GPU memory usage; Develop automated functional, accuracy, and performance regression tests.
Seniority
Senior, hands-on IC
