Software Engineer Intern, Inference

What you'll do

• Profile and benchmark model inference; identify latency, throughput, and memory bottlenecks.
• Optimize inference, data pipelines, and GPU kernels, and build performance benchmarks and tooling.
• Work with researchers to implement infrastructure improvements and present project results.

What they ask for

• Currently pursuing a BS, MS, or PhD in Computer Science, Electrical/Computer Engineering, or a related field.
• Strong programming skills in Python and/or C++.
• Solid understanding of deep learning fundamentals and at least one framework; PyTorch preferred.
• Comfortable analyzing performance data and reasoning about latency, throughput, memory, and accuracy trade-offs.

Nice to have

• Experience with video model inference, systems or ML research, latency-critical inference, multi-GPU inference, or open-source ML systems projects.
• Experience optimizing inference, data pipelines, or GPU kernels using relevant tools and techniques.

Tools

Python, C++, PyTorch, CUDA, Triton, TensorRT, ONNX Runtime, TVM

AI focus

Optimize inference for large video and world models, including quantization, distillation, sampling acceleration, attention and caching, compilation, and GPU kernel optimization.