Member of Technical Staff - AI Data Platform, Frontier Models
Core
Building large-scale AI data infrastructure and intelligent data pipelines to support pre-training and post-training of frontier multimodal models.
Role type
Senior IC AI Data Platform Engineer (Frontier Models)
Builds
EB-scale data platforms, intelligent multimodal data processing systems, and AI-native data pipelines for model training.
Domain
Artificial Intelligence, Large Language Models (LLMs), Multimodal AI, Data Engineering
Deliverable
production ML models
Required skills
Python, SQL, Shell, Distributed Data Platforms (Spark, Flink, Ray), Multimodal Data handling, LLM/VLM pre/post-training data construction, Synthetic data generation, Evaluation benchmarks (MMLU, MMBench), Open file/table formats (Lance, Iceberg, Paimon, Parquet), LLMs/Vision models understanding, ETL systems, Data modeling, Data pipelines
Preferred skills
Large-scale text/image/video dataset processing, Agents and modern LLM toolchains, Data governance, Hybrid search, Rare high-value dataset discovery
Technologies
Spark, Flink, Ray, Lance, Iceberg, Paimon, Parquet, Python, SQL, Shell, LLMs, VLMs
Responsibilities
Develop and evolve large-scale AI data infrastructure for frontier AI labs; Build automated understanding and processing systems for text, images, and video; Design and train models for data classification, recognition, and quality scoring; Leverage LLMs and Agents to build intelligent data pipelines for automation; Design open, AI-optimized storage layers supporting multimodal data and embeddings; Develop differentiated datasets for challenging domains using real-world collection and synthetic generation; Drive the Data-Model-Evaluation iteration loop to optimize model capabilities.