Senior Applied Scientist, Efficient LLM Inference & Model Optimization

Kurz zusammengefasst

The Senior Applied Scientist role focuses on building fast, high‑throughput inference pipelines for large language models at a rapidly expanding, globally distributed AI company. You will work closely with machine‑learning engineers, product groups and hardware experts to embed quantisation, pruning and compiler‑level tricks into production services, shaping the company's AI infrastructure and influencing product direction. Day‑to‑day you will design, implement and benchmark low‑latency inference stacks, integrate frameworks such as PyTorch, TensorRT, ONNX or TVM, and use Python and C++ to turn research advances into reliable code. The job also involves GPU/TPU acceleration, distributed inference setups and performance profiling with tools like Nsight or Perf, while contributing to open‑source efficient‑AI projects. The position requires a Ph.D. or equivalent experience in machine learning, deep‑learning systems or computer architecture, plus a proven record of deploying LLMs at scale and strong Python/C++ skills. Experience with the listed frameworks, hardware acceleration and profiling tools is essential; familiarity with open‑source contributions is a plus. What the role asks for: - Ph.D. or equivalent in ML, deep‑learning systems, or computer architecture - Proven large‑language‑model deployment at scale - Strong Python programming - Strong C++ programming - Experience with PyTorch, TensorRT, ONNX or TVM - GPU/TPU acceleration experience - Distributed inference experience - Performance profiling using Nsight or Perf (nice‑to‑have)