LinkedIn is seeking a Staff Software Engineer to join their AI Platform team, focusing on scaling large model training and serving infrastructure. The role involves working with cutting-edge AI technologies including LLMs, recommendation systems, and computer vision models. The team is responsible for optimizing performance across algorithms, frameworks, and hardware, managing thousands of GPU cards.
The position spans multiple critical areas: Model Training Infrastructure, where you'll build next-gen training systems for AI use cases, work with popular libraries like Huggingface and PyTorch, and enable distributed training for massive models. Feature Engineering, where you'll develop the Feature Platform handling millions of QPS and terabytes of data. Model Serving Infrastructure, focusing on low-latency applications serving large complex models. And MLOps, covering metadata, observability, orchestration, and experimentation systems.
This is an opportunity to advance one of the world's most scalable AI platforms while working with talented researchers and engineers. The role offers significant technical challenges in distributed systems, deep learning optimization, and large-scale data processing. You'll have the chance to influence open-source projects and define the future of AI infrastructure at LinkedIn.
The position combines technical leadership with hands-on engineering, requiring both strategic thinking and deep technical expertise. You'll mentor other engineers while staying close to cutting-edge AI developments. The hybrid work environment offers flexibility while maintaining strong team collaboration and culture.