Senior Research Data Engineer
Core
Build data pipelines, datasets, and tooling to fuel multimodal agent research and training loops.
Role type
Senior Research Data Engineer (Multimodal AI)
Builds
Scalable training and evaluation loops for multimodal agentic systems
Domain
AI Research / Multimodal Modeling / LLM/VLM Training
Deliverable
production ML models
Required skills
Python, ML DevOps, distributed systems, data versioning, cloud infrastructure (AWS), prompt engineering, data quality monitoring, synthetic data generation, human-in-the-loop workflows
Preferred skills
RLHF/DPO training, preference data collection, multimodal data handling (image-text, video), benchmark creation
Technologies
Python, Ray, S3, AWS, Label Studio
Responsibilities
Design and build data pipelines for agent training (collection, filtering, deduplication, formatting, versioning); Maintain infrastructure for efficient data loading, storage, and retrieval at scale; Collaborate with researchers to translate requirements into data specifications; Create evaluation datasets and benchmarks; Develop tooling for dataset construction including annotation workflows and synthetic data generation; Own data quality through validation frameworks and drift monitoring; Document datasets thoroughly with provenance and versioning history; Implement test coverage for data pipelines and ML workflows; Contribute to roadmaps by identifying data bottlenecks
Seniority
Senior, hands-on IC