How Bloomberg's Data Scientists use Satellite Images for Reporting: Krishna Karra - MBM78
Krishna Karra is a data scientist & report for Bloomberg, having used machine learning & satellite images for reporting. Recent stories from him & his team include mapping refugee camps in Rafah & exposing illegal ship oil transfers in the middle of the Ocean.Sponsor: Beemaps by HivemapperGet access to high quality, fresh map data at https://beemaps.com/mindsUse promo code MINDS to get 50% off your API credits through Dec. 31 2024About KrishnaTwitterLinkedInShownotesNote: Links to books are Amazon Affiliate links. I earn a small commission if you buy any of these books.Bloomberg: The Clandestine Oil Shipping Hub Funneling Iranian Crude to ChinaBloomberg: A Detailed Map Shows How Airstrikes and Refugees Reshaped RafahHow Radar Satellites See through Clouds (Synthetic Aperture Radar Explained)National Land Cover Database (NLCD)What Ukraine Has LostGraves in Suda by Joe MorrisonJean Martin Bauer on Minds Behind MapsBooks & Podcast:Overstory by Richard Powers (Affiliate Link)Ezra Klein ShowTimestamps(00:00) - Intro(00:34) - Sponsor: Beemaps(01:51) - Krishna describes himself(03:27) - Example stories: Illegal Oil transfers(05:29) - Stories are the goal(07:07) - Why publish the data set?(12:24) - How Journalism has and hasn't changed(14:04) - How data changes a story(18:23) - Putting the datasets together(20:37) - Conveying trust(24:07) - Showing the limitations of the data(26:11) - Why is journalism important for satellite data?(30:14) - News room process(32:57) - Building custom tools(38:19) - Timeline of a news story(39:47) - What Krishna has learned as a data scientist in a news room(40:49) - Stories that have stuck out(42:57) - Different ways of showing the data(44:19) - Krishna's wishlist(51:12) - Book & podcast recommendation(53:16) - Paid podcasts & media(55:19) - Support the podcast on PatreonSupport the podcast on PatreonMy TwitterPodcast TwitterRead Previous Issues of the NewsletterEdited by Peter XiongFind more of his work
--------
57:03
How Ex-Vox Video Producer Phil Edwards thinks about Maps to tell stories - MBM77
Phil Edwards is a video producer who worked at Vox for nearly 10 years, and now runs his own Youtube channel exploring the history of businesses, and lately has been using more and more maps. We go over one of his latest videos, “The Secret Economics of Google Street View” as a case study of how Phil thinks about maps to tell stories. We also talk about journalism on Youtube, and the business behind running a Youtube channel today.Sponsor: Beemaps by HivemapperGet access to high quality, fresh map data at https://beemaps.com/mindsUse promo code MINDS to get 50% off your API credits through Dec. 31 2024About PhilTwitterPhil's BlueskyPhil’s YoutubeShownotesNote: Links to books are Amazon Affiliate links. I earn a small commission if you buy any of these books.Phil’s trivia siteGeoGuessrPhil’s video about Google Street ViewBook & Podcast recommendations:3 Scientists & Their Gods by Robert Wright (Affiliate Link)The Evolution of God by Robert Wright (Affiliate Link)On the Edge by Nate Silver (Affiliate Link)Freakonomics podcastAcquired PodcastMy conversation with Ashlee VanceWild Wild Space documentaryNeoHoogTimestamps(00:00) - Intro(00:52) - Sponsor: Beemaps(02:08) - Who is Phil Edwards?(03:13) - What does a Video Producer do?(04:29) - Chosing topics & stories(06:38) - Maps in video production(10:14) - Bringing a different element to maps stories(12:15) - Explaining Complicated Things(15:49) - Case Study: Google Street View Video(20:07) - Geoguessr(23:03) - Story & Journalism on Youtube(27:25) - Stories that Phil wants to work on(30:54) - Importance of visuals(34:58) - Learning Curve in Mapping & Animation(40:26) - Balancing Scientific Rigour and Story(49:53) - The Business of Youtube(54:09) - Choosing to scale or not(59:14) - Creating "content"(01:01:56) - Authenticity(01:05:00) - Valuing High Quality Journalism(01:10:18) - Succeeding on Patreon(01:13:46) - Creators that Phil admires(01:16:56) - Books & podcasts Recommendations(01:28:54) - Reporting On Vs Working In a field(01:31:40) - Where Phil finds inspiration(01:35:14) - Support the podcast on PatreonSupport the podcast on PatreonMy TwitterMy BlueskyRead Previous Issues of the NewsletterEdited by Peter XiongFind more of his work
--------
1:36:23
What Hyperspectral constellations have to offer: Pixxel’s example - Awais Ahmed #MBM76
Awais Ahmed is the co-founder & CEO of Pixxel, a company building a constellation of hyperspectral imaging satellites. Unlike “traditional” cameras, these satellites can see across hundreds of bands, opening up a lot more applications. We talk about the engineering -and funding- required to pull this off and how Awais manages a company between India (where Awais is from) and the US. But that’s just the 1st part of the interview. Then we talk about his vision for the company, which goes way, way beyond sending imaging satellites.Sponsor: OpenCageUse OpenCage for your geocoding needs with their APIGeomobAbout AwaisTwitterLinkedInShownotesAwais’s personnal websitePixxel’s 3 phase visionMaxar HD machine learning upscalingBook & Podcast RecommendationThe Beginning of Infinity by David Deutsch (Affiliate Link)Profiles of the Future by Arthur C. Clarke (only found it in ebook here, Affiliate Link)Hitchhiker’s guide to the Galaxy by Douglas Adams (Affiliate Link)Timestamps(00:00) - Introduction(01:09) - Sponsor: OpenCage(02:34) - Awais describes himself(03:46) - Pixxel(05:07) - What is hyperspectral?(10:36) - Spacial, Temporal & Spectral Resolution tradeoffs(18:01) - Hardest part of building satellites(22:58) - The resolution spotlight(28:38) - Image Quality Rabbit Hole(30:42) - The tricky part of raising money(38:52) - Awais has a big, big vision(50:42) - Likeliness of Awais's vision within 10 years(54:14) - Working between India & USA(59:04) - Personnal investing(01:01:20) - Importance of Reading(01:05:36) - Sci-Fi & History(01:08:33) - Book recommendation(01:10:43) - Awais in space?(01:11:54) - Pushing the boundaries of humanity(01:13:14) - Optimism in a cynical world(01:19:11) - Support the podcast on PatreonSupport the podcast on PatreonMy video on an introduction to satellite imagesWebsiteMy TwitterPodcast TwitterRead Previous Issues of the NewsletterEdited by Peter XiongFind more of his work
--------
1:20:55
The Ex-Uber Data Scientist Who wants to simplify Data Science with Serverless Computing: Sina Kashuk
Sina Kashuk is the co-founder & CEO of Fused, who wants to make iterating & deploying in Python faster with serverless computing. We break down what that actually means, why it matters and what data science workflows could look like over the next few years.This also isn’t Sina’s first company, a few years ago he started Unfolded.ai, focused on making visualisations for data scientists faster. The company was acquired by Foursquare in 2021.Sponsor: Beemaps by HivemapperGet access to high quality, fresh map data at https://beemaps.com/mindsUse promo code MINDS to get 50% off your API credits through Dec. 31 2024About SinaTwitterLinkedInShownotesNote: Links to books are Amazon Affiliate links. I earn a small commission if you buy any of these books.My blogpost joining the teamUber’s H3 tiling gridFoursquare acquires UnfoldedAWS LambdaMy conversation with Ib GreenFused.ioBook & Podcast recommendationAwaken the Giant Within by Tony Robbins Affiliate LinkI’m pretty sure you can find Minds Behind Maps by yourself if you’re hereTimestamps(00:00) - Intro (02:38) - Sponsor: Beemaps(03:55) - Hacking (06:07) - Fused.io(07:23) - Why run your algorithm in the cloud? (10:06) - Serverless computing (12:40) - Optimizing for iteration speed (18:52) - Breaking Fused into smaller parts (23:27) - "User Defined Functions: UDF" (31:08) - How do you make money? (31:56) - Why start companies? (42:41) - Convincing people to use your tools (49:44) - Speed isn't all: Train / Plane analogy (54:36) - Going beyond geospatial (57:33) - Building a team (59:54) - Podcast/book recommendation (01:01:11) - Building a Long Term Vision (01:06:59) - Support the podcast on PatreonSupport the podcast on PatreonMy TwitterPodcast TwitterRead Previous Issues of the NewsletterEdited by Peter XiongFind more of his work
--------
1:08:43
Meet the man trying to build a 20cm 3D map of the world: Andrew Peterson - #MBM74
Andrew Peterson is the Co-Founder & CEO of Array Labs, with a simple mission: Mapping the whole world in 3D, at 20cm in near real time.We peel the layers as to what it takes to get there: the engineering that’s required, how to build a constellation to do that, how you fund such a project.Sponsor: OpenCageUse OpenCage for your geocoding needs with their APIGeomobAbout AndrewLinkedInArray LabsShownotesNote: Links to books are Amazon Affiliate links. I earn a small commission if you buy any of these books.My 20min video explaining radar satellite imagesFind USGS 1m LiDAR data across the USY CombinatorMy episode with Ashlee VanceMy episode with Brian McClendonAlbedoBook & Podcast recommendationSpin Selling by Neil Rackham (Affiliate Link)AcquiredTimestamps(00:00) - Intro(01:08) - Sponsor: OpenCage(02:34) - "Being a Recovering Engineer"(03:15) - Mapping the world in 3D(09:59) - "Near Real Time"(15:46) - Applications will only use what's available, by definition(18:15) - Why use radar for 3D images?(22:23) - The coolest Space Shuttle mission, period(27:19) - Tradeoff between resolution & coverage(36:26) - Building cheap radar satellites(39:46) - Array Labs's image resolution(45:10) - A GPU Analogy(50:34) - A story of image processing & computers(56:07) - Array Labs today(57:57) - Let's talk $$$(01:06:38) - Low barrier to entry: Comparing XRay & MRI(01:12:09) - Why stop at 10 satellites?(01:15:50) - Focus(01:19:53) - Max & Andrew's 1st chat during covid(01:23:22) - Subscription model for satellite images?(01:32:50) - Convincing the rest of the world your idea is worth something(01:43:58) - Engineer to Founder(01:47:30) - Book & Podcast recommendation(01:51:06) - Array Labs's next 4 years?(01:53:45) - Support the podcast on PatreonSupport the podcast on PatreonMy video on an introduction to satellite imagesWebsiteMy TwitterPodcast TwitterRead Previous Issues of the NewsletterEdited by Peter XiongFind more of his work