About this role
Build an end-to-end data system for sourcing, labeling, processing, synthesis, and evaluation to continuously produce high-quality datasets and evaluation sets for foundation model development; responsibilities include designing scalable data infrastructure, multimodal training data strategies, building a data–model–evaluation closed loop with Agents, and applying large-model data techniques such as augmentation and distillation in business scenarios.
Skills for this role
PythonJavaSQLData structures and algorithmsSparkFlinkRayLarge modelsMultimodal AIAIGCLLMDiffusionT2VCLIPVLMAgentsTool useAgent trajectoriesAgent memoryMulti-turn interactionData storageData ingestionData cleaningData labelingData managementData analysisData synthesisData processingData augmentationData distillation」「Data filtering