Hamed Alemohammad: Open Machine Learning for Satellite Imagery - MBM#34

Hamed Alemohammad was the previous Executive Director of the Radiant Earth Foundation, a Non-Profit focused on building open machine learning tools & datasets for satellite imagery through projects like the ML Hub & STAC, Spatio Temporal Asset Catalog. 

--- 
Episode Sponsor: OpenCage
Use OpenCage for your geocoding needs with their API. 
They have a generous Free trial you can sign up to! 
---

About Hamed 
- Twitter
- LinkedIn

Shownotes

• LandCoverNet
  • Paper
  • ML Hub datasets
• Radiant Earth Foundation
• Data Driven
• Clark names inaugural director of Clark Labs: Center for Geospatial Analytics
• Book Recommendation
  • Atomic Habits by James Clear

Timestamps 
00:00 Introduction 
03:01 Hamed's Interest for Earth Observation 
08:53 Joining the Radiant Earth Foundation 
10:25 What is the Radiant Earth Foundation? 
12:14 Why a Non-Profit? 
15:34 Finances & Incentives: Where does the money come from? 
19:10 Choosing to focus on Machine Learning; ML Hub 
26:43 Capacity & Community Building 
30:42 Only a 10 people team 
34:44 Comparing to Google Earth Engine, AWS & Microsoft Planetary Computer 
41:01 Making Decisions & Measuring Progress in a Non-Profit 
44:12 Machine Learning Competitions & Kaggle 
48:22 Challenges of Working with External Organizations 
51:19 Building Models is Sexy but only a small part of Machine Learning 
55:55 Who is the ML Hub build for? 
57:59 From Researcher to Executive Director 
01:01:37 What does being on the Board of Directors Mean? 
01:06:18 Hamed's Journey from Iran to working in the US 
01:11:03 Advice for people outside the US & Europe 
01:15:46 The limitations of English for a truly worldwide reach

Feel free to reach out! 
- Website
- My Twitter
- Podcast Twitter
- Read Previous Issues of the Newsletter
- Edited by Peter Xiong. Find more of his work