Hyphen Connect is seeking an AI Specialist Engineer to enhance the performance of large language and vision models for on-device inference. The role involves developing and deploying cutting-edge AI solutions to ensure optimal efficiency across diverse hardware architectures.
Responsibilities:
- Compress and optimize large language and vision models for on-device inference
- Develop pipelines for model distillation and hardware-specific compilation
- Benchmark performance across various NPU/GPU architectures
Requirements:
- Expertise in model distillation, pruning, and 4-bit/8-bit quantization techniques
- Hands-on experience with TensorRT, ONNX Runtime, and edge deployment
- Strong C++ and Python skills