Machine Learning Infrastructure Engineer, Recommendations and Search at TikTok USDS Joint Venture · San Jose, CA · Joined
TV
Machine Learning Infrastructure Engineer, Recommendations and Search
TikTok USDS Joint Venture · San Jose, CA · Just now · 0 applicants
NewOn-siteFull-timeMiddle
Pay
Pay not listed
Workplace
On-site
Job type
Full-time
Level
Middle
Team: Recommendations and Search · Source: jobright · Posted by joined-scrapper
About the role
Build and scale distributed training and inference platforms for TikTok's recommendation and search engines, focusing on large-scale parallelism, production serving, and secure compliant systems.
What you’ll do
Build and scale distributed training and inference platforms
Optimize large-scale parallelism for recommendation and search engines
Ensure production model serving and infrastructure resiliency
Maintain secure and compliant compute and data systems
Optimize high-performance training loops and eliminate bottlenecks
Collaborate across teams, write technical design docs, and mentor team members
What you’ll need
Bachelor's or Master's in Computer Science or related technical discipline
2+ years of software engineering experience with Python, C++/Java
1+ years of direct experience building and maintaining ML infrastructure at enterprise scale
Deep familiarity with PyTorch/TensorFlow, CUDA, and networking topologies
Experience scaling LLM training/inference across hundreds of GPUs
Experience with advanced serving techniques like KV Cache, PD separation, and quantization
Original job description
TikTok USDS Joint Venture is hiring a Machine Learning Infrastructure Engineer to build and scale distributed training and inference platforms for TikTok’s recommendation and search engines. The role focuses on large-scale parallelism, production model serving, infrastructure resiliency, and secure, compliant compute and data systems.Bachelor's or…