>4x improvement on geospatial tasks with map in the loop.
The graph shows a baseline 2% task success rate improving to to 8% task success rate, but the evals section details 100% success rates across the board.
I'm not sure what the effectiveness of this skill is from the readme. Is it 8% success, or 100% success?
Question from an outsider: Who is paying for tools like this? The examples shown on the website (e.g. all streets in Nevada) look nice, but what are those analyses actually used for? I am pretty sure it is not only about having pretty maps but their has to be a business value I don’t see right now.
20 year GIS dev here. Looks pretty useful for data exploration. I'd say one of the more compelling GeoAI things I've seen.
The problem is there's really a lot of data out there and it's a lot of work to move it around, e.g. between S3 buckets. There's also a ton of GIS SAAS vendors who are pure rent-seekers: I'm looking at a newer offering charging $23 per month for 10GB storage. This has more utility than their offering in my opinion.
The good thing here is that it could keep data provenance because it's SQL over known datasets.
Unrelated, but as someone who is on the verge of also creating another GIS offering do you think there is any value to creating a low cost hosting platform centered around data portability? This came out of frustration with the existing landscape of offerings and I put together something that I wish existed.
I work with maps everyday. I'm cheap and my employer is cheap with me, but we've got to produce a lot of maps for compliance & business intelligence. The work is is mostly cleaning & standardization, with some user experience toward a particular audit purpose.
There are some much more lucrative niches, that have to do with chain-of-title, rights of way, resource rights, and so on, and I can imagine why anyone would pay to save, say, 20 hours a week.
Power interconnects for datacenter siting would be a hot example.
This can be very useful for urban planning. you could have an agent investigate the optimal spot for a new datacenter, examine solar power installations, and so on.
Exactly, this platform has fallen down so incredibly low. Every other post is worthless garbage about LLMs, without a single ounce of actual science being showcased, created, or even talked about. But a whole post about a markdown file is a new low imo. How does anyone who's actually competent at all in their domain think that this is worth sharing?
See: https://github.com/ignfab/geocontext (French) Beta MCP instance: https://geollm.beta.ign.fr/geocontext/mcp
Unrelated, but also take a look at the nice high-density LiDAR point data we have! https://visionneuse-lidarhd.ign.fr/?px=4441970.281583222&py=...
The graph shows a baseline 2% task success rate improving to to 8% task success rate, but the evals section details 100% success rates across the board.
I'm not sure what the effectiveness of this skill is from the readme. Is it 8% success, or 100% success?
The problem is there's really a lot of data out there and it's a lot of work to move it around, e.g. between S3 buckets. There's also a ton of GIS SAAS vendors who are pure rent-seekers: I'm looking at a newer offering charging $23 per month for 10GB storage. This has more utility than their offering in my opinion.
The good thing here is that it could keep data provenance because it's SQL over known datasets.
There are some much more lucrative niches, that have to do with chain-of-title, rights of way, resource rights, and so on, and I can imagine why anyone would pay to save, say, 20 hours a week.
Power interconnects for datacenter siting would be a hot example.
Either LLMs will be so good in a few months this will be redundant.
Or it won't be and LLMs are a dead end and there are better ways to build with LLMs