Senior Applied Scientist
Core
Research and optimize efficient inference for frontier LLMs and VLMs, translating research findings into production capabilities for Nebius's full-stack AI cloud platform.
Role type
Senior Applied Scientist (Efficient Inference & Systems)
Builds
Production-ready inference engines, optimized serving architectures, and distributed inference systems for LLMs/VLMs.
Domain
AI/ML Systems, Large Language Models, Distributed Computing
Deliverable
production ML models
Required skills
LLM/VLM inference optimization, quantization, model distillation, speculative decoding, KV-cache management, distributed inference architecture, experimental design, PyTorch, CUDA, Triton
Preferred skills
NeurIPS/ICML/MLSys publications, vLLM/SGLang/TensorRT-LLM experience, post-training optimization (SFT/DPO/RLHF), wide expert parallelism (WideEP)
Technologies
PyTorch, Triton, CUDA, vLLM, SGLang, TensorRT-LLM, FlashAttention, NVIDIA Dynamo
Responsibilities
Lead research projects in efficient LLM/VLM inference from hypothesis to production handoff; Develop and evaluate methods spanning quantization, distillation, and speculative decoding; Build prototypes using CUDA-adjacent tooling and inference-serving frameworks; Research LLM request routing, scheduling, and multi-node inference strategies; Develop rigorous evaluations covering latency, throughput, and cost per token; Collaborate with engineering teams to integrate research into production components.
Seniority
Senior, hands-on IC
