智能模型数据平台工程师-AI Data
Core
Build a unified platform for the full lifecycle management, production, consumption, and quality assurance of massive datasets for the Seed large language model.
Role type
Senior IC AI Data Platform Engineer
Builds
Unified data platform for LLM pre-training, SFT, and RLHF; Data Agents; High-throughput data infrastructure (data lakes, object storage, distributed computing, vector databases).
Domain
Artificial Intelligence / Large Language Models / Data Infrastructure
Deliverable
production ML models
Required skills
Python, Java, Go, Spark, Flink, Ray, Hudi, Iceberg, Paimon, Doris, ClickHouse, PyTorch, TensorFlow, Indexing and Search Architecture
Preferred skills
LLM data scraping, processing, and synthesis for pre-training/SFT/RLHF; Agent project development; Harness Engineering practices
Technologies
Spark, Flink, Ray, Hudi, Iceberg, Paimon, Doris, ClickHouse, PyTorch, TensorFlow, Vector Databases, Data Lakehouse
Responsibilities
Design and implement data pipelines for PB-scale daily data processing; Collaborate with infrastructure teams to build data lakes and distributed computing platforms; Lead technology selection and architecture optimization for high fault-tolerance systems; Develop Data Agents to reconstruct AI-Native data platform paradigms.
Seniority
Senior, hands-on IC