@dggoldst.bsky.social
Partner Research Manager at Microsoft Research NYC. AI, Economics, Decision Science.
On the other hand if you dreamed it that's good news because now you can publish it
10.10.2025 23:07 β π 1 π 0 π¬ 0 π 0By @rfoucart.bsky.social
10.10.2025 23:06 β π 38 π 4 π¬ 2 π 0Unless it was all a dream
10.10.2025 22:58 β π 1 π 0 π¬ 1 π 0I don't know. Here's AI's attempt
chatgpt.com/s/t_68e977d8...
What does it mean to say you figured something out by yourself?
We benefit from inheriting thousands of years of advances in mathematical notation, tools, and methods.
5,000 years ago nobody on earth would be able to figure out problems that today's schoolchildren can "figure out by themselves."
when kids are trying to understand something they use ai in smarter ways than we give them credit for
10.10.2025 11:36 β π 11 π 1 π¬ 1 π 0That would have been so funny if I saw you there at the info booth!
10.10.2025 02:25 β π 0 π 0 π¬ 0 π 0Ah, now I see! That must be it!
10.10.2025 01:09 β π 1 π 0 π¬ 1 π 0According to economist Thomas Schelling, if 2 strangers had to meet in NYC without planning, many would choose the info booth at Grand Central at 12:00 PM on January 1.
At 8:38 PM on October 9th it was just me and the info booth guy.
π¨ Postdoctoral Research Economist positions at Microsoft Research New England or New York City π¨
www.aeaweb.org/joe/listing....
Open to PhDs in econ-related fields (e.g. quant marketing, econ-CS). Seeking those with high collaboration potential with the team: www.microsoft.com/en-us/resear...
If you get addicted there is a harmless variant called Synonym Toast Crunch
06.10.2025 12:43 β π 4 π 1 π¬ 1 π 1Screenshot of Tweet by @dggoldst Oct 12, 2021 "I hereby preregister that Susan Athey will win a Nobel Prize in Economics"
Repeating this prediction (that Susan Athey will win the Nobel) until I'm right.
06.10.2025 11:47 β π 18 π 1 π¬ 1 π 1Peanut butter contains no peas, no nuts, and no butter
05.10.2025 20:28 β π 10 π 0 π¬ 3 π 0chatGPT 5
chatgpt.com/s/t_68e28bba...
It's funny. When things are phrased as math problems LLMs are in the mathematical elite. And then there's this. I wonder if it's the smaller cheaper models used to supplement search results at scale without breaking the bank? Or just that it doesn't have the trappings of a math problem.
05.10.2025 15:14 β π 1 π 0 π¬ 1 π 04 and 1/3?? Is this a front door or a bank vault?
05.10.2025 00:59 β π 2 π 0 π¬ 0 π 01 and 3/4??
05.10.2025 00:53 β π 1 π 0 π¬ 1 π 0Just curious did you use Gen AI to help solve the problem?
After I read your answer I asked it.
π€£
04.10.2025 12:12 β π 1 π 0 π¬ 0 π 0You might ruin the door but if you were going to trash it anyway...
04.10.2025 03:09 β π 1 π 0 π¬ 0 π 0Before replacing the door I would be tempted to take a router and cut a mortise 5/16" deep on each face of the door in exactly the shape of the smart lock housing. Then your door will be 2 1/8" just where the lock goes. If the door is hollow you could glue support blocks inside the door.
04.10.2025 03:08 β π 1 π 0 π¬ 2 π 0Everyone thinking about becoming a professor should read this
nymag.com/intelligence...
Readers, it is important to note Andrea's capitalization here.
03.10.2025 20:54 β π 0 π 0 π¬ 0 π 0It's a publicity still from the play The Blot starring Brad Pitt
29.09.2025 22:56 β π 2 π 0 π¬ 1 π 0Someone should make a data science textbook / cookbook where you do the lesson and then eat the meal cooked by the laptop.
29.09.2025 16:00 β π 3 π 0 π¬ 3 π 0This was about 20 minutes before sunset. The desk lamp soon became necessary :)
29.09.2025 02:20 β π 5 π 0 π¬ 0 π 0Electric grill with corn and onions on a urban balcony with a desk light
Having a good ol' fashioned Manhattan BBQ
28.09.2025 22:50 β π 36 π 1 π¬ 5 π 0We've all been there... π«
27.09.2025 21:03 β π 15 π 4 π¬ 0 π 0Ha!
27.09.2025 20:09 β π 2 π 0 π¬ 0 π 0