Frederik Kratzert
Research Scientist @ Google Research. Working on the intersection of machine learning and hydrology. Located in Vienna, Austria.
- Reposted by Frederik KratzertI present the most interesting graph ever made. HUMAN ON BICYCLE beats every other living thing. www.scientificamerican.com/article/a-hu...
- Interesting read and something I have been thinking about recently. How fit is our education system for the new era and what does this mean for _learning_ (especially learning to learn), if students/pupils over rely on LLMs. What does this mean for my kids (2nd grade and kindergarten)...?
- Some updates on NeuralHydrology and I'll also use this opportunity to give a shout-out to @gauchm.bsky.social for his continuous support and maintenance of NH
- Reposted by Frederik Kratzert[Not loaded yet]
- Reposted by Frederik Kratzert[Not loaded yet]
- We (wife, 7 y/o, myself) spent New Years Eve playing "Stuffed Fables" boardgamegeek.com/boardgame/23... My wife's first roleplay game and also the most complex game we played so far with our oldest. Time was flying and we only made it to the start of the second chapter 😅
- Reposted by Frederik Kratzert[Not loaded yet]
- Interesting point. I've been thinking about this as well in the past and I think, especially with ML, we have the opportunity to optimize for what we really want to know from the model, instead of training it on some proxy task
- Reposted by Frederik Kratzert[Not loaded yet]
- Reposted by Frederik Kratzert[Not loaded yet]
- Kid #2 got sick in time for Christmas and it seems I caught whatever he has, too. For the last three years we had at least one kid sick at Christmas. Happy Holidays 🎄
- Reposted by Frederik Kratzert[Not loaded yet]
- It is so wild how all academics seem to have decided that LinkedIn is the new platform. Essentially the same announcement receives 100x the engagement.. On a job market platform. In the good old days of Twitter, this was different and it's kinda sad that not more researchers decided to join BlueSky
- Excited to unveil Caravan-Qual! 🌊 We've expanded the Caravan framework to bridge the gap between water quantity and quality. This new open dataset features: 🔹 ~70M observations 🔹137k stations 🔹 100 constituents 🔹 Linked streamflow, forcing & attributes Links in 🧵
- 📦 Full Data: doi.org/10.24416/UU0... 📂 Lite (CSV): zenodo.org/records/1778... 💻 Code: github.com/SustainableW... 📖Preprint: eartharxiv.org/repository/v... #OpenScience #Hydrology #WaterQuality
- First time in years that I won't attend #AGU Inspiring week for everyone who attends and I hope you all stay safe.
- This looks amazing 😲 Deserved first bookmark on bluesky
- Reposted by Frederik Kratzert[Not loaded yet]
- Quick shout-out for an #EGU26 session that I'm not involved in but think is worth sharing: EOS4.4 BUGS: Blunders, Unexpected Glitches, and Surprises For a detailed session description, see www.egu26.eu/session/56997
- What do people do when their taxi driver is a) speeding like crazy and b) interacting with his phone all the time while driving. Not sure if I'm unlucky but this happens 4/5 times to me when taking the airport taxi back home...
- It has been a while since the last news from Caravan but I'm happy to announce that a new extension has just been published. Caravan-CZ, a dataset curated by CHMI itself in cooperation with Grey Nearing. For details check: github.com/kratzert/Car... All extensions: github.com/kratzert/Car...
- The paper is finally published and we were able to keep the original title. hess.copernicus.org/articles/29/...
- jax.config.update("jax_debug_nans", True) saved my day.
- Oh my god, I finally was able to fix it. After effectively not using my private laptop for the past 9 months (I have a corp laptop and a private desktop PC that runs Linux), I sat down and ask Gemini for help. I know, I know... Thread for details.
- This has easily been one of the most interesting external meetings in a while.
- Reposted by Frederik Kratzert[Not loaded yet]
- Reposted by Frederik Kratzert[Not loaded yet]
- Looking forward to a full day of exciting talks at the GRIT workshop in Oxford.
- I just learned that the recording of my invited plenary keynote talk at #EMS25 last week in Ljubljana, Slovenia is publicly available. Here is the link in case anyone wants to watch it: vimeo.com/1118021056?f...
- After about one year in review, the dataset paper for the Caravan-GRDC extension was finally published. See: essd.copernicus.org/articles/17/... I am so happy that GRDC was open for this effort and made some of their data freely available through Caravan.
- Reposted by Frederik Kratzert[Not loaded yet]
- A packed room full of young and motivated scientists, today at my presentation at the @ellis.eu #SummerSchool2025. It was a real pleasure being there. Now I'm on my way back to Vienna but not without the by-now normal delay of 1h+ with the @bahn.de...
- Never managed to attend a summer school during my master or PhD program. But it is never too late =) Arrived in Jena to attend "ELLIS Summer School: AI for Earth and Climate Sciences" and to give a talk later this week. Looking forward to many exciting conversations with young scientists
- @simonwillison.net have you seen that?
- Reposted by Frederik Kratzert[Not loaded yet]
- Reposted by Frederik Kratzert[Not loaded yet]
- Oh, I didn't know. That is indeed pretty neat.
- Das kann doch kein zufall sein @geschichte.fm ?!
- We are in the year 2025 and people write a paper showing that a model trained with ‼️global, coarse resolution forcings‼️ performs worse in ‼️ungauged‼️ basins, than a regional model trained with ‼️local, high resolution forcings‼️ in ‼️gauged‼️ basins. This is used to say "training on more basins is bad"🤯
- You know this feeling, when you get to review a paper for the n-th amount of time (always resubmissions to different journals after rejections), with the authors not adapting any of the comments and keeping all the obvious non-sense? Nobody ever should see my first draft of review comments.....
- Reposted by Frederik Kratzert[Not loaded yet]
- I wish my universities would have offered such classes. Sounds amazing
- @gauchm.bsky.social and @danklotz.bsky.social you are not alone
- Trying to install Ruby and RVM on Linux for 15 minutes and wondering if there is really something more complicated than setting up Python.
- Reposted by Frederik Kratzert[Not loaded yet]