Route optimization API that solves fleet planning as a single problem
Dynamic geography clustering beats fixed zones for large-scale routing problems.
APL-inspired symbolic system that makes LLMs invent new glyphs for your problems.
AI researchers, prompt engineers, cognitive science enthusiasts
If you read the core module it may seem insane in some places, and it should because there are 'hallucinated' concepts and conclusions (not defending them). Regardless, it's very fun to watch them invent new symbols to help you see the structure of whatever problem you are working on. It is what it is, and it works anyway. I find that interesting.
If you are familiar with the APL programming language you may see where this is going - https://computerhistory.org/blog/the-apl-programming-languag....
You can find the claude code plugin on github - https://github.com/xyzcoordinate/TopoGlyph. (will add more examples, haven't taken the time to organize things)
Recently Anthropic's Fable model was banned (idk the right term). It would be annoying if a similar thing happened to this. I wrote the whole text onto the Ethereum blockchain and locked it.
- https://github.com/xyzcoordinate/TopoGlyph/blob/main/topogly... - read it with foundry: `cast call 0xe1888706EC9C071E149060aeF70B539eBeb61092 "profileOf(uint256)(string)" 2 --rpc-url https://ethereum-rpc.publicnode.com ` - https://etherscan.io/address/0xe1888706EC9C071E149060aeF70B5...
That is all. Have a look if you are curious, is pretty awesome.
Dynamic geography clustering beats fixed zones for large-scale routing problems.
Unverified LLM attempt at open math problem without expert validation.
Alpine meets React in three symbols, but Alpine already exists and works.
Whitepaper arguing for scientific knowledge routing infrastructure without any actual software.
Photomath for physics with a Socratic tutor mode, but the category is already crowded.
Problem curation is solid; execution is a README with 4 ideas and zero traction so far.