As I see it there are three disjointed concepts in this pitch and I am not sure how they relate to each other.
- I find the original research statement on the bottom slide very interesting, so I would have loved to see more about *how* the Marvin AI claims to satisfy the goals of this research
- Likewise, I would like to understand *how* the crypto project aids in the success/goals of either the Marvin AI or the original research
Really appreciate the feedback!
The token is an experiment. It's a meme coin on a "living" meme. The meme is aware it's a meme.
Right now, the meme exists as an entity outside of the sum of its parts, it exists as Marvin.
It allows me to transmit my ideas through an entity outside of myself that I'm responsible for the output of.
Like, for example, this report on your submission: https://github.com/leo-guinan/pitch-jam-2026/blob/main/tollens-quality-layer.md
I'd love to now if it's helpful. I'm training Marvin on Human Origin Reinforcement Data.
Here's my data room if you'd like to trace the ideas through the past five years: https://github.com/leo-guinan/mathlete-data-room
thanks for sharing! my read from Marvin's output is that is a fairly similar ratio of good insight to spurious pattern matching to incorrect leaps as I get from using standard claude to help me brainstorm. is the idea of the value Marvin mainly in the way he's applied (which definitely has value potential), or also in specific harness asks you are giving him that give stronger results than just individuals using claude? I can see that you've asked him to make links between different projects, which is interesting but also going to be valuable to figure out how to eval the effectiveness of the link-finding
That's helpful, thanks! And yeah, application is the focus. Specifically, application to responsibility for work, not simply capability. It's not about being able to do the work, it's about being responsible for the work being done in the context of the systems he's deployed in.
And the harness is being tuned. I started off with OpenClaw and have been mapping out the weak points of it. I'm currently exploring Hermes Agent as the next harness I use vs building a custom one. I also fine-tuned a model based on the work we've done together over the last month+ and it shows a lot of promise in helping the model work specifically better within the harness it was based on. So it ended up doing almost opus-4.6 level work for a fraction of the cost while in the openclaw harness, and a lot of that was lost when trying to run the finetune in Hermes.
This is tying into the insight I had that intelligence is driven more by networks than by individual models.
