Trajectory, a startup building a post-training platform for continual learning, has raised a $45 million Series A led by Sequoia at a $300 million valuation. Nvidia, Bessemer Venture Partners and Bain Capital Ventures also participated, and Trajectory opened its self-serve platform to the public on November 2.
“Imagine if everyone had the tools to shape intelligence itself. That day is today.”— Ronak Malde, Trajectory co-founder and CEO
Trajectory was founded about a year ago by three Stanford classmates and has 18 employees. Co-founder and CEO Ronak Malde was among the researchers Google DeepMind acqui-hired from Windsurf. Co-founder Michael Elabd previously worked at DeepMind, and co-founder Arjun Karanam worked on generative foundation models at Apple. The company has also hired researchers and product designers from OpenAI, Stripe and Netflix.
The product targets the step after pre-training. Businesses start with an open-source model, connect it to their applications, and collect feedback on how it performs. Trajectory then uses that feedback to retrain the model, so the model improves from production use. The company says the platform works with any agent harness and needs two code changes to get started.
Key facts
- 01Trajectory raised a $45M Series A led by Sequoia, valuing the company at $300M.
- 02Nvidia, Bessemer Venture Partners and Bain Capital Ventures also joined the round.
- 03The startup was founded about a year ago by three Stanford classmates and has 18 employees.
- 04Its self-serve post-training platform launched publicly on November 2, 2026.
- 05Harvey and Rogo use it to post-train open-source models for legal and financial work; Decagon, Mercor and Nvidia are also customers.
Malde's pitch is that general-purpose models know a lot but lack job-specific experience. He compared foundation models to "a very smart PhD student who's always starting a new job."
“But what the economy values is someone who worked on the job for 30 years,”— Ronak Malde, Trajectory co-founder and CEO
With the platform, users "decide what good looks like for their organization and can turn that into improvements for their model," Malde said. Trajectory is releasing two methods as research previews: Continual Grading, which captures expert judgment as a training signal, and Group Size 1, which trains on single production rollouts.
Harvey and Rogo use Trajectory to post-train open-source models for legal and financial tasks. Decagon, Mercor and Nvidia are also customers. Trajectory previously published a field note on post-training Nvidia's Nemotron 3 Ultra on Harvey LAB in under 24 hours.
Trajectory's website lists $60 million raised across its seed and Series A. Backers shown there include Sequoia, Conviction, Nvidia, Notion, Hugging Face, Dropbox, Braintrust and Modal, along with Fei-Fei Li and Jeff Dean.
The startup enters a crowded post-training market. Competitors include Thinking Machines Lab, maker of the Tinker fine-tuning platform; Applied Compute, which was valued at $3.25 billion last month; and Prime Intellect.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
