How to Build a Search Engine for Geography
Searching for places is still surprisingly primitive compared to text, images, or products.
Spot Scout is a search engine built specifically for geography.
It makes the physical world searchable using maps, satellite imagery, and terrain data.
Things like elevation, slope, visibility, land cover, distance, and access are modeled computationally.
Vague queries like “remote but reachable” or “a good sunset spot near a road” turn into concrete spatial operations.
The talk gives a behind-the-scenes look at the tech and what this could enable next.
Context:
I love looking at maps, especially when I am planning to travel and want to find spots for myself. For example, if I am in Gokarna and want to watch the sunset, where should I actually go? Ideally somewhere a bit elevated, with a clear view, and not too far from a road.
So far, doing this has been a very manual process. I look at terrain maps, figure out which side a hill faces, check forest cover to see if I can even stand there, and then try to guess if the spot will work. What struck me was that even in the age of AI, this kind of search is still very primitive.
That led to this idea. What if finding places like this was simple? What if you could just describe what you want, and the system figures out the geography for you. That is what pushed me to build something to make finding spots effortless for people who enjoy exploring.
About the speaker:
Jayesh enjoys exploring new technologies and putting them to work on real-world problems. Alongside tech, he spends a lot of time looking at maps, reading about foreign policy, playing sports, and hiking whenever possible. He likes building things driven by curiosity rather than trends, and enjoys speaking in public and learning from people with very different experiences.
https://x.com/wjayesh
To attend online:
Add to calendar: https://bit.ly/45Mb4bC
Gmeet link: https://meet.google.com/zth-mzqe-fxe?hs=122&authuser=0