Imagine if everyone had the tools to shape intelligence itself. That day is today.
Today we launch the Trajectory platform - the easiest and most intuitive tool to post-train models of your own. The foundation for continual learning.
With just a few agent prompts, you get access to a full creative suite to train, eval, and deploy trillion parameter models.
On top of that, we believe in a future where you don’t just shape your model once, it also keeps learning on its own from the world around it.
We’re releasing two research previews.
1. Continual grading scales your taste, so your model learns what good looks like from far more than you could ever grade by hand.
2. GS1, a new algorithm that lets models learn directly from what they do in production.
We’ve now raised 60M, with a new round led by @Sequoia at a 300M valuation to democratize the most powerful technology of today - training models - so that the lawyers, doctors, and engineers of the world can own their own intelligence.
You know what good looks like, now go build it. Own your own value function with Trajectory.
Quick update -
We’ve been letting 100 people off the waitlist every day, but signups are still growing faster than that.
In the meantime, we’ve been bringing our top users over to the Trajectory office in SF to meet the team.
We’re working hard to speed things up so you can start training on the Trajectory platform today!
The goal of our research is to enable AI to learn from experience in the real world just like humans do, while keeping human's value function in the loop.
This is the ultimate alignment problem, and we make two significant strides in achieving this goal, so that AI can learn online, and in the directions that matter to you.
One is a new method, and the other is a new algorithm. See below.
Models have IQ but no experience.
Frontier models show up on day one with superhuman intelligence but no expertise on the thousands of tasks your company actually does. @trajectorylabs exists to close that gap through continual learning.
@rronak_ trained SWE-1 at Windsurf, then went to DeepMind. His cofounders @MichaelElabd and @QuantumArjun worked on Gemini Robotics and multimodal models for Vision Pro. All three hit the same wall: big labs train one model for everyone, but nobody trains one for you.
The conversation covers:
— why today’s agents are “pseudo continual learning”: we are in the paper road map era, not Google Maps
— “Group Size 1”, a GRPO variant that learns from one real-world rollout instead of replaying the same task eight times, so you don’t need a simulated environment
— how Harvey trained a model 6x cheaper than frontier that took an open-source baseline up by over 4x on its own legal benchmark
— Continual Grading, a meta-grader that turns rubrics on 100 traces into reward for 10,000
— continual learning vs. recursive self-improvement, and why only one keeps humans in the loop
— why continual learning will make product-building more like tending a garden of products
Chapters
0:00 – Introduction
0:45 – Pseudo continual learning: paper maps vs Google Maps
2:48 – Do we need new architectures? Rich Sutton, David Silver, and the missing layer
4:33 – From a $100M consulting model to a product: lessons from SWE-1 at Windsurf
7:36 – Harvey: 6x cheaper than frontier, 4% to 18% on its own benchmark
8:55 – Why "Trajectory": the unit of work is a task, not a token
10:13 – The primitives of continual learning: traces, replayable environments, reward
12:56 – Group Size 1 and the meta-grader: training without an environment or a labeled reward
16:16 – The platform: train, eval and deploy in one click, and who it's for
18:58 – Rogo and intelligence density
20:14 – What's next: curiosity, and why continual learning beats RSI on alignment
24:19 – Research and product under one roof, and the product builder as gardener
27:26 – Why work at Trajectory, and the optimistic case: from 1% to 99%
Was an honor to go on Sequoia’s podcast!
We talk about how current models are like we back are in the days of AAA roadtrips, where you go to an agency to plan out every road ahead of time.
A platform for post-training gets you to at least MapQuest, where everyone gets that ability to plan their own trip.
What true continual learning unlocks though is Google Maps, where your route is updated realtime and modifiable at every turn
We will see that same transformation with model training as we democratize the capability to shape intelligence realtime
Trajectory CEO @rronak_ explains how "vibe RL" could let anyone train their own personalized AI model using just a few prompts:
"I want people to vibe RL basically. You can train a model to get better at F1 racing. You can train it to be good at chess or sudoku, or just learn how to sound like you when you're texting."
"Everyone has data about themselves. They don't want it going over to the frontier labs, but you want some sort of personalized intelligence that knows so much about you."
"Right now it is literally just a couple agent prompts to get going. We have a waitlist open right now. Very soon I want it to be every single person in the world can be post-training a model."
@trajectorylabs
We've been taking companies off the waitlist as fast as we can since our launch yesterday!
In the meantime, here's an inside look into the product itself, and how you can train models with just a few agent prompts
We’re proud to be backed by some incredible folks who have helped us along the way - Sequoia, Conviction, NVIDIA, Bessemer, BoxGroup, Bain Capital, SV Angel, Exceptional Cap, Creandum, WndrCo, Genius Ventures, Pear VC, Afore VC, Base10, Designer Fund, First Harmonic, Standard VC, Vine Ventures, and many more!