Continual Learning: How AI Agents Get Better With Every Use | Arjun Karanam, Trajectory
Trajectory is building a continual learning platform to close AI agents' 'experience gap'
“These models are smarter and smarter but always feels like when you're talking to them, it's their first day on the job.”
Trajectory co-founders argue that while foundation models are rapidly improving in raw capability (IQ), they lack accumulated experience — every agent run generates tokens that encode real work and decisions, but that signal is discarded. Their platform aims to capture and learn from those agent outputs so AI systems improve with use, analogous to how human expertise compounds over time.