Machine Learning Engineer - Model Optimization & Inference

About the opportunity

Optimize deep learning models for high-throughput GPU inference and test accelerated AI pipelines.

What you will do

• check_circle Benchmark model inference speed and memory footprint across modern GPU architectures.
• check_circle Apply quantization (FP16, INT8) and pruning techniques to vision and language models.
• check_circle Containerize deployment pipelines using Docker, Triton Inference Server, and Kubernetes.
• check_circle Collaborate with application engineers to debug pipeline bottlenecks.

What we are looking for

• arrow_circle_right Solid programming skills in Python and foundational knowledge of C++ or CUDA concepts.
• arrow_circle_right Hands-on experience with PyTorch or TensorFlow model development and export (ONNX).
• arrow_circle_right Understanding of GPU computing, memory bandwidth, and neural network latency trade-offs.
• arrow_circle_right Passion for hardware-software co-optimization.

Skills & Tech Stack