Senior AI Training & Serving Infra Engineer (Hybrid)

LinkedIn; California (MO); Hybrid; USD 144,000 - 236,000; Full time

Job summary

LinkedIn is seeking a Senior Software Engineer to scale its AI Training Infra, supporting large models and feature engineering at scale. You will design high-performance data I/O, enable distributed training across hundreds of billions of parameters, and work with PyTorch, TensorFlow, CUDA, and other DL stacks.

You will also build containerized pipelines, improve observability, and collaborate with research teams to advance ML serving, MLOps, and experimentation at LinkedIn.

Qualifications

• Bachelor's degree in CS or related field or equivalent practical experience.
• 2+ years building deep learning systems and distributed software.
• Experience with Java, C++, Python, Go, or similar languages and distributed systems.
• Strong knowledge of ML infra, GPUs, and modern DL frameworks.

Responsibilities

• Design, implement, and optimize large-scale distributed serving/training for recommendations and LLMs.
• Improve observability and developer productivity across systems.
• Mentor engineers and contribute to a fast-growing AI team.
• Collaborate with open-source communities (PyTorch, Huggingface, etc.)

Skills

Data Structures & Algorithms

Backend Systems Infrastructure

ML Algorithm Development

Machine Learning and Deep Learning

Information Retrieval, Recommendation,

Education

Bachelor's Degree in Computer Science or related technical discipline

BS and 5+ years or MS and 4+ years or PhD and 2+ years

Tools

PyTorch

TensorFlow

CUDA

cuDNN

NCCL

TorchRec

Flint

Spark

Beam

Job description

LinkedIn is seeking a Senior Software Engineer to scale its AI Training Infra, supporting large models and feature engineering at scale. You will design high-performance data I/O, enable distributed training across hundreds of billions of parameters, and work with PyTorch, TensorFlow, CUDA, and other DL stacks.

You will also build containerized pipelines, improve observability, and collaborate with research teams to advance ML serving, MLOps, and experimentation at LinkedIn.