Description
Eventual is hiring a Research Engineer on its Visual Understanding team to own the roadmap for making petabytes of video queryable by content. The role involves selecting, training, fine-tuning, and evaluating vision-language models, VQA models, embeddings, and CNNs; reducing annotation costs through distillation and pipeline optimization; building queryable datasets; and partnering with data, storage, and partner-research teams. The position is applied research focused on production impact, requires experience with vision and multimodal models at scale, and is based four days per week in the San Francisco Mission District office.
