Daniel Rothenberg

@danielrothenberg.com

Tech-evangelizing Atmospheric Scientist | co-founder @brightbandtech | Formerly Tech Lead @Waymo, Chief Scientist @tomorrowio_, @eapsMIT

How good are the frontier AI weather forecast models from Google DeepMind, Microsoft, and NVIDIA? See for yourself on @brightband.com's Operational WeatherBench dashboard - owb.brightband.com! We're happy to run these models operationally and share our evaluation stats with the community!

Operational WeatherBench | AI Weather Forecast Leaderboard

Live skill rankings for AI and physics-based weather forecast models, scored daily with standard metrics and an open toolkit.

owb.brightband.com

Are folks in mainstream climate science aware of the large and growing number of VC-backed tech startups pursuing all manners of geoengineering approaches? The genie seems to be completely outside the bottle at this point.

Is it just me or is 90% of social media content (particularly on TikTok) about student loans? Like... I don't have student loans. No reason I should be fed this content through organic engagement. Gotta say, there's some _discontent_ out there right now...

Friendly reminder for colleagues at Harvard facing disappointing NSF news today - there's a pretty good law school across campus with an impressive alumni network. I seem to recall them also facing the ire of the current Administration. Perchance a coffee chat with a colleague to align incentives?

I think I should start charging for prospecting calls. I respect the hustle for folks selling services to startups, but time is my single most valuable resource. If I politely reply that I don't need or want your service, please respect that - or be prepared to compensate me.

I think the hardest challenge of working independently / leading an organization is being satisfied that you've "done enough" on any given day. It's like the early PhD phase on steroids - there's always so much to do, and priorities aren't always obvious!

Might need a CAPTCHA for future economics discussions with people: If on Monday the market loses 10% of its value, and on Tuesday it recovers 10% during the course of the day, then what is the net impact on the market as of Wednesday morning?

Had a really cool idea for a new NCAAM win probability forecast, similar to 538's, but using my own ELO-based model. Unfortunately there's a showtime bug and I doubt I'll have time to get it running this evening :( Maybe next year!