AI Weekly Malaysia

Drew Edwards' Flatten SF is a browser-based tool that finds flat routes between any two points in San Francisco by computing over 160,000 street segments with USGS 1 m lidar elevation and the Overture/OpenStreetMap street graph. A slider exposes the trade-off between shortest and flattest routes, treating a foot of climb as costing 200 ft of walking and using cumulative elevation gain; stairways are allowed on foot but excluded for bikes, and place search is built in offline. The Hacker News discussion has 162 points and 54 comments.

Why: For builders, the reusable piece is the client-side architecture and the explicit distance-vs-climb trade-off, not the SF-specific dataset. If you work on maps, routing, or local utilities, study how 160,000 segments and offline place search run in-browser; Malaysian builders would need to check whether equivalent open elevation and street data exist before adapting it to hilly local areas.