Machine Learning Software Engineer (Modelzoo)

About the role

NXP is looking for a Machine Learning Software Engineer (2-3 years of experience) to work on its Modelzoo team in Hyderabad. The role focuses on porting and optimizing deep learning models for specialized hardware, with a particular emphasis on quantization to improve inference performance.

Key responsibilities

• Port and deploy deep learning models from frameworks such as PyTorch and TensorFlow to proprietary or commercial ML accelerator hardware platforms.
• Analyze and improve model performance for the target hardware, focusing on latency and throughput.
• Contribute to model quantization efforts (e.g., INT8) to reduce model size and speed up inference while keeping accuracy.
• Use profiling tools to find and fix performance bottlenecks in the ML inference pipeline on the accelerator.

Required profile

• Bachelor's or Master's degree in Computer Science, Electrical Engineering or a related field.
• 2-3 years of experience.
• Hands-on experience deploying and optimizing models on GPUs or other specialized accelerators.
• Some experience with post-training quantization.

Required skills

• Deep learning frameworks: PyTorch and TensorFlow.
• C++ and Python.
• GPU programming models such as CUDA/cuDNN (a plus).
• ML inference engines and runtimes such as TensorRT, OpenVINO and TensorFlow Lite.
• Git and collaborative development workflows.
• Preferred: hardware-aware model design, compiler technologies for deep learning, real-time or embedded systems, cloud platforms (AWS, GCP, Azure), and CI/CD pipelines for ML models.