Trajectory launches continual learning platform after $60M raise

Trajectory has launched a new platform designed to let companies create custom versions of open-weight models trained on real product usage and continual feedback. Co-founded by Ronak Malde, Michael Elabd, and Arjun Karanam, the company raised $60 million in funding with participation from Sequoia, NVIDIA, and Designer Fund. The platform captures production traces and user feedback to improve model performance on specific tasks while keeping product teams in control of training and updates. Open article →