Trajectory, a startup founded by alumni from DeepMind, Apple, and OpenAI, secured $15 million in seed funding at a $115 million valuation. The company focuses on developing continual learning models that improve through user interactions.
Trajectory is tackling a fundamental challenge in AI development: building models that learn and adapt continuously from real-world user data rather than remaining static after initial training.
The startup was founded by veterans from three of the industry's most influential organizations. This pedigree signals confidence in the team's ability to execute on a complex technical problem that has eluded broader implementation.
The Continual Learning Approach
Unlike traditional AI models that are trained once and deployed, Trajectory's vision involves systems that improve through ongoing interactions. This approach mirrors how humans learn—accumulating knowledge and refining behavior based on experience.
The company believes rapid iteration cycles can accelerate AI product development across industries. By automating the process of learning from user data, enterprises could deploy AI solutions that become more capable over time without constant retraining from scratch.
Market Opportunity
The $115 million valuation reflects investor confidence in the market need. Companies across sectors—from customer service to productivity tools—struggle with AI systems that fail to adapt to new information or changing user needs after deployment.
Trajectory's approach could address this gap by enabling continuous improvement without requiring data scientists to retrain models manually. The company is positioning itself as infrastructure for companies building AI products rather than an end-user application.
Next Steps
With $15 million in funding, Trajectory can expand its team and develop its core technology. The seed round suggests investors see viability in the continual learning model as a differentiator in an increasingly crowded AI landscape.
The startup joins a wave of AI infrastructure companies attempting to make model development faster and more efficient. Success would require proving that continual learning delivers practical advantages over existing approaches and can scale across diverse use cases.
Town, an enterprise personal AI assistant platform, is in talks to raise funding at a $1 billion valuation in a round led by Index Ventures, according to sources.
Product management platform Linear reached $100M in annual recurring revenue and conducted a $99M employee tender offer at a $2.5B valuation, doubling its valuation from $1.25B in 2025.
Flipboard is acquiring Bluesky feed-building startup Graze, integrating its ad technology and creator monetization tools into its open social web platform.
Artificial intelligence is fundamentally changing how startups operate, not just what they build. According to Floodgate co-founder Ann Miura-Ko, companies are deploying AI agents across engineering, sales, marketing, and strategy to accelerate learning and execution.