About this role
Design, execute, and operationalize fine-tuning workflows for large language models across supervised, preference-based, and reinforcement learning approaches, and operate complex training pipelines with rigorous evaluation and dataset construction. Work cross-functionally with product, design, engineering, operations, and business stakeholders, and provide code/design review and mentorship.
Skills for this role
PythonPyTorchMachine LearningLarge Language Models (LLMs)Fine-tuning transformer-based language modelsDistributed trainingFSDPZeROPipeline parallelismRLHFDPOEvaluation methodologyHuman evaluation designGPU clustersDataset constructionSynthetic data generationDataset distillationMultimodal model fine-tuningResponsible AI evaluationRed-teamingCommunication skills