Senior Director, Applied Research

PhD in Computer Science, Machine Learning, Computer Engineering, Applied Mathematics, or Electrical Engineering. Experience with Large Language Models (LLMs), including training models from scratch (10B+ parameters, 500B+ tokens), publications at ACL, NAACL, EMNLP, Neurips, ICML, or ICLR, working on open source or commercial LLMs, guiding large-scale model training teams, experience with 500+ node GPU clusters, and knowledge of training optimization frameworks like DeepSpeed or NeMo. Expertise in Behavioral Models (Geometric Deep Learning, Graph Neural Networks, Sequential Models, Multivariate Time Series), including technical leadership for large user behavior models, publications at KDD, ICML, NeurIPS, or ICLR, scaling graph models to 50m+ nodes, experience with recommender systems, production real-time/streaming environments, contributions to open source frameworks like PyTorch Geometric or DGL, and experience with datasets of 100m+ users. Specialization in Optimization (Model Sparsification, Quantization, Training Parallelism, Gradient Checkpointing, Model Compression) with 5+ years of experience or publications. Knowledge of Finetuning (Supervised, Instruction, Dialogue, Parameter Tuning), transfer learning, model adaptation, and deploying fine-tuned LLMs. Publications on tokenization, data quality, dataset curation, or labeling, leading contributions to large open source corpora (1 Trillion+ tokens), or core contributions to open source libraries for data quality and labeling.