lmeierhoefer

12 ms·

1.
by lmeierhoefer 1y ago
We are not doing RLHF but fine-tuning directly on a reward function. Our task was around improving a coding agent, coding in JSONata( https://jsonata.org ). GPT4o is quite bad in this, as there are not too many JSONata snippets on
2.
by lmeierhoefer 1y ago
Yes, we wanted to incentivize, that people who want to use the platform (redeeming the $20 training credits) are also joining a slack channel, so we can give direct support. We should have pointed this out in the post.
3.
by lmeierhoefer 1y ago
No, not really. As I posted in the other thread, there are quite a few historical examples of why the big labs won’t take the entire market. They will push to publish something like this soon. Also, I think reinforcement fine-tuning is more
4.
by lmeierhoefer 1y ago
Thanks! Yes, absolutely. OpenAI already has a reinforcement learning fine-tuning API in closed beta. However, historically, they’ve always left significant room for integrations into users systems. E.g. in the current demo of their RL fine-
5.
by lmeierhoefer 1y ago
Yes, great point. We are currently working on multistep RL. The big problem with the trivial approach (give a single reward to the entire (ReAct) trajectory) is that the model receives a weak learning signal per decision (called credit assi
6.
Launch HN: Augento (YC W25) – Fine-tune your agents with reinforcement learning
101 points by lmeierhoefer 1y ago|60 comments
7.
Show HN: Open Source Reinforcement Fine-Tuning for Your Agents(augento.ai)
5 points by lmeierhoefer 1y ago|1 comments
8.
by lmeierhoefer 1y ago
Hi HN, we are building a reinforcement-learning fine-tuning service for LLMs. As we know, agents fail all the time. Especially when you try to use them for something actually useful. Current solution approaches suck; prompting has intrinsic
9.
by lmeierhoefer 2y ago
You are right that, at the moment, the system inherently requires a 64-bit OS. We currently support Debian-based distros; it should work with other parent distributions as well, but you need to translate the installer script ;) But we defin
10.
by lmeierhoefer 2y ago
We will most likely go with an open-core model. The main part will stay open source (the Core OS extension is under GPL3, and everything SDK-related is MIT). For paid features, we have several ideas: a hosted management plane to configure a
11.
by lmeierhoefer 2y ago
Thanks for the feedback! > I would just take a centralized framework and stick it inside an oversized drone/vehicle capable of carrying the added weight Makes sense. I think there are scenarios where such “base stations” are a prior
12.
by lmeierhoefer 2y ago
> Is this primarily a passion project or are you hoping to get corporate sponsorship & adoption? We are in the current YC W25 batch and our vision is to build a developer framework for autonomous robotics systems from the system we a
13.
by lmeierhoefer 2y ago
Yeah, SLAM seems also like a natural showcase for us. I am just working on a decentralised collaborative SLAM package on top of our system, where multiple robots can drive around and continuously merge their maps without a coordination serv