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.
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.
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.
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.
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?
>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?
Map snapshot PNG. Apparently, LLM is quite competent when reading map images. It can say, “Oh, that's not all London coverage.” “ “Oh, this and this street is a problem (without having street data).”
I am currently working on a website https://hillsha.de that makes it easy to download LiDAR las/laz files for almost every place in europe, the US and some other regions. I also made an iOS app for the same use-case, which can render the LiDAR data in 3D and 2D without PDAL and GDAL. It uses a vibe-coded library instead that combines both in native Swift. The iOS app is still in testing but works great.
Implementing France was a lot more comfortable than almost every other country, very well structured metadata and naming conventions. So thanks for that
(i work at the german mapping agency but this is a private project since i just love working with LiDAR hillshades)
Thanks for sharing, and thanks for making the video you shared. A couple thoughts. Is PostGIS king in this area? I've been liking duckDB as it does not require a server. Is there a good alternative to PostGIS in the duckDB world? Maybe just plain GeoParquet files read in through duckDB?
In the example in your video are any special GEOMETRY functions being used in the underlying SQL? Or, could your data just have been in plain postgres?
A similar product GeoSQL is Malloy which puts a semantic layer on top of your data for better LLM understanding. Malloyyo gives you an MCP server for precise and auditable interaction with your data.
23 comments
[ 3.0 ms ] story [ 52.8 ms ] threadThe 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
That said, some of the skill frameworks like gstack created 10x productivity gain for me. IMO, worth sharing here.
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?
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=...
I am currently working on a website https://hillsha.de that makes it easy to download LiDAR las/laz files for almost every place in europe, the US and some other regions. I also made an iOS app for the same use-case, which can render the LiDAR data in 3D and 2D without PDAL and GDAL. It uses a vibe-coded library instead that combines both in native Swift. The iOS app is still in testing but works great.
Implementing France was a lot more comfortable than almost every other country, very well structured metadata and naming conventions. So thanks for that
(i work at the german mapping agency but this is a private project since i just love working with LiDAR hillshades)
In the example in your video are any special GEOMETRY functions being used in the underlying SQL? Or, could your data just have been in plain postgres?
A similar product GeoSQL is Malloy which puts a semantic layer on top of your data for better LLM understanding. Malloyyo gives you an MCP server for precise and auditable interaction with your data.