3: Predicting Crop Yields, Satellite Imagery and Machine Learning with David Potere of TellusLabs


Manage episode 185967434 series 1536774
By Mike Betts. Discovered by Player FM and our community — copyright is owned by the publisher, not Player FM, and audio is streamed directly from their servers. Hit the Subscribe button to track updates in Player FM, or paste the feed URL into other podcast apps.

David Potere is Co-founder and CEO of TellusLabs, a company that combines decades of satellite imagery with machine learning to predict crop yields and help its customers, which range from farmer to hedge fund manager.

Get the full show notes at here.

If you like the show, you can subscribe to it on any podcast app. Support us by leaving a review on iTunes, Apple Podcasts, or Stitcher Radio. Reviews really help the show grow and allow other people find the show, it only takes a minute.

We cover:

  • [02:55] Humble beginnings in satellite imagery from the Navy to Boston Consulting Group.
  • [09:34] The range of problems TellusLabs is solving for its customers.
  • [17:01] TellusLabs products Kernel vs Argus
  • [19:01] Why satellite imagery now.
  • [22:43] Training models with decades of satellite imagery data, discussion of imagery quality.
  • [29:49] USDA approach to forecasting and how TellusLabs differs.
  • [33:40] How approach to yield forecasting in U.S. differs from other countries.
  • [35:11] Yield forecasting in Brazil and LatAm.
  • [40:36] More on how customers are using these solutions, from hedge funds to farmers.
  • [51:52] Market size vs market potential for satellite imagery solutions.
  • [55:36] Lessons learned building TellusLabs.
  • [58:13] Boston startup ecosystem, distributed teams, startup accelerators.

8 episodes