Trajectory raises $45M to make AI model training accessible to nontechnical experts
The one-year-old startup is now valued at $300 million.
TL;DR
- Current AI models require additional training to learn from mistakes.
- Businesses often fine-tune AI models using expensive forward-deployed engineers (FDEs).
- Trajectory aims to automate AI model improvement through continual learning.
- The startup raised $45 million in a Series A round at a $300 million valuation.
- Trajectory's platform allows businesses to connect AI models, collect feedback, and retrain them for specialized tasks.
- This can make deploying specialized AI faster and cheaper for professionals like lawyers and doctors.
- Companies like Harvey and Rogo are using Trajectory's platform for post-training open source models.
- Trajectory's approach emphasizes human involvement in the training process to ensure safety.
- Competitors include Thinking Machines Lab, Applied Compute, and Prime Intellect.
- Trajectory's advantage lies in its team's expertise and its user-friendly interface for nontechnical customers.