Staff Software Engineer - AI Platform
Core
Building and scaling LinkedIn's AI platform for model training, feature engineering, and serving large-scale models (LLMs, recommendation, computer vision) with hundreds of billions of parameters.
Role type
Staff Software Engineer (AI Platform Infrastructure)
Builds
Next-gen training infrastructure, high-performance feature platforms, low-latency model serving applications, and MLOps/experimentation systems.
Domain
AI/ML Infrastructure, Distributed Systems, Large-Scale Data Processing
Deliverable
production ML models
Required skills
Distributed systems design, Deep learning frameworks (PyTorch, TensorFlow), GPU optimization, Feature engineering, Container orchestration, Open-source contribution, System observability
Preferred skills
LLM serving, Search systems, ML applications, Distributed data processing (Spark, Flink, Beam), Container technologies, Deep learning tensor libraries
Technologies
PyTorch, TensorFlow, Huggingface, DeepSpeed, vLLM, Spark, Beam, Flink, Kubernetes, CUDA, RDMA, NCCL, MLFlow, Kubeflow
Responsibilities
Design and implement high-performance distributed training and serving infrastructure; Optimize GPU fleet utilization and deep learning frameworks; Build and maintain feature platforms for online/offline environments; Lead technical strategy for AI infrastructure initiatives; Mentor engineers and define technical culture; Collaborate with open-source communities.
Seniority
Staff, strategic technical leadership & hands-on IC