多模态大模型数据工程师-产品研发
Core
Design and develop large-scale pre-training data processing pipelines and platforms for foundation models (LLM, VLM), including data sourcing, scraping, parsing, lifecycle management, and synthetic data generation.
Role type
Senior IC multimodal large model data engineer
Builds
Stable, reliable data processing pipelines and platforms for base model pre-training
Domain
Artificial Intelligence / Large Language Models / Multimodal AI
Deliverable
production ML models
Required skills
Python, Go, Java, Spark, Flink, Kafka, Hive, HDFS, data pipeline architecture, data governance, synthetic data generation frameworks
Preferred skills
Data middle platform development, machine learning system platform development, deep understanding of LLM/VLM ecosystem, data experimentation engineering
Responsibilities
Design and develop data processing pipelines for model pre-training; Build data platforms for metadata, lineage, and storage governance; Develop data synthesis solutions and frameworks for scaling; Abstract and develop efficient data processing frameworks for algorithm engineers