4 months ago
Bengaluru, IndiaStaff+
Responsibilities
- Lead a team developing and enhancing model converter code for PyTorch and Keras trained models.
- Collaborate with machine-learning modeling and embedded software teams to convert trained models for Syntiant Neural Decision Processors.
- Optimize code and extract maximum performance from the processors when generating neural-network models from pretrained models.
- Investigate and optimize embedded firmware and the deep-neural-network execution path as needed.
- Work across the model-converter and embedded-firmware stack to support edge AI devices.
Requirements
- At least 8 years of experience developing ML frameworks or applications for edge devices.
- Strong experience with C, C++, and Python.
- Experience with ML frameworks such as TensorFlow and PyTorch.
- Experience with ML runtimes such as TVM, ONNX, TF Lite Micro, or TensorRT.
- Experience with embedded development, bare-metal systems, and RTOS.
- Experience with code optimization, quantization, and network pruning.
- Experience working in large, complex codebases.
Benefits
- Full-time, on-site position in Bengaluru, India.
- Base pay is part of the total compensation package, with individual decisions based on experience, skills, location, schedule, and peer compensation; no specific salary range is stated.
Categories
