Research Scientist, Distributed Data & Model State

TikTok Inc.; San Jose (CA); On-site; USD 162,000 - 388,000; Full time

Job summary

TikTok Inc. is seeking graduates to join the Arch-Data Ecosystem team within the TikTok Recommendation System.

You will work on offline and real-time data storage solutions handling large-scale data for recommendations, search, and advertising, serving over 1 billion users. As a new hire, you will contribute to building scalable Lakehouse architectures, integrate training data and inference state systems, and collaborate with ML platforms to enable PyTorch model training and efficient data

Qualifications

• PhD in a relevant technical discipline or closely related field.
• Experience building large-scale distributed systems (storage, streaming, ML infra).
• Understanding of Lakehouse technologies (e.g., Flink + Paimon, Iceberg, Delta Lake or Hudi).

Responsibilities

• Design real-time and offline data architectures for large-scale recommendation systems.
• Build scalable streaming Lakehouse systems powering feature pipelines and model training.
• Collaborate with ML platform teams to support PyTorch-based training workflows and data formats for large-scale samples.
• Own components of the distributed storage and processing stack, including file formats and metadata management.

Skills

PhD in related field

Distributed systems

Apache Flink

Lakehouse technologies

Education

PhD in Software Development, CS, or related

Tools

Apache Flink

Paimon

Iceberg

Delta Lake

Job description

TikTok Inc. is seeking graduates to join the Arch-Data Ecosystem team within the TikTok Recommendation System.

You will work on offline and real-time data storage solutions handling large-scale data for recommendations, search, and advertising, serving over 1 billion users. As a new hire, you will contribute to building scalable Lakehouse architectures, integrate training data and inference state systems, and collaborate with ML platforms to enable PyTorch model training and efficient data