I started a podcast with my friend Krishna Karra! Krishna is extremely knowledgeable, and I firmly believe that he is one of the people who has looked at the most earth observation pixels on the planet.
We worked together briefly in 2022, on a satellite imagery retrieval tool called Earth Index. Throughout this time, we spent a lot of time investigating the current topic du jour: embeddings. Since interest is increasing in this technology, we thought we’d share our opinions and experiences working with them and developing applications on top of them.
Our first episode is titled: “Embeddings in Context: Tools for Geospatial Problem Solving” and you can find it on spotify, apple and youtube.
Give it a listen to find out:
what problems we solved use embeddings
how I spent $40,000 in cloud credits on pre-training on satellite imagery only to be beaten by ImageNet weights
why I think embeddings for search should be high recall across a broad range of problems at the expensive of precision, and more!
I’ll be writing about some of these topics in more depth too, you can subscribe below so you don’t miss those pieces.
Nice podcast! Nice throwback to my NM days. I don't think I noticed when we were at Descartes that Krishna had a sexy radio voice. Anyway, the podcast got me thinking about some things, so a mini-monograph is coming your way by email.