» teej (@teej_m) 's Twitter Profile
» teej

@teej_m

Building @_downlink – I make AI 3x faster

ID: 17101063

calendar_today01-11-2008 14:51:48

12,12K Tweet

9,9K Followers

1,1K Following

Sterling Crispin 🕊️ (@sterlingcrispin) 's Twitter Profile Photo

This is one of the craziest ideas I've ever seen. He converted a drawing of a bird into a spectrogram (PNG -> Soundwave) then played it to a Starling who sung it back reproducing the PNG. Using the birds brain as a hard drive with 2mbps read write speed. youtube.com/watch?si=HMtVd…

This is one of the craziest ideas I've ever seen. He converted a drawing of a bird into a spectrogram (PNG -> Soundwave) then played it to a Starling who sung it back reproducing the PNG.

Using the birds brain as a hard drive with 2mbps read write speed. 
youtube.com/watch?si=HMtVd…
Jacob Matson (@matsonj) 's Twitter Profile Photo

built baby's first eval: I want to find the best LLM for co-writing. What that means for me is I want it to have a great way of thinking in patterns of words, not numbers. so using NYT Connections puzzles seemed to be the perfect subject.

built baby's first eval: I want to find the best LLM for co-writing. What that means for me is I want it to have a great way of thinking in patterns of words, not numbers. 

so using NYT Connections puzzles seemed to be the perfect subject.
ben (@benhylak) 's Twitter Profile Photo

This is insane: GPT-5 supports context-free grammars You can define your DSL and constrain the output to it. The alternative was having some model-compatible JSON representation that was converted into your DSL. So many little things adding up here.

This is insane: GPT-5 supports context-free grammars

You can define your DSL and constrain the output to it. 

The alternative was having some model-compatible JSON representation that was converted into your DSL.

So many little things adding up here.
Jo Kristian Bergum (@jobergum) 's Twitter Profile Photo

We need more 200-lines examples of building an agent loop with tools using the primitive APIs. We need to demystify agent implementations. We don’t need another agentic framework.

» teej (@teej_m) 's Twitter Profile Photo

I worked for a social app with millions of users. We spent half of eng capacity on bookmark folders. The app was dead within 18 months. Cultural relevance is the only priority.

dex (@dexhorthy) 's Twitter Profile Photo

Who’s working on integrating brainrot into my terminal so I can resist the temptation to spin up a 5th parallel Claude while the 4th one is working

» teej (@teej_m) 's Twitter Profile Photo

People want AI to solve their problems. Math and code are hard, verifiable problems. Labs are laying the groundwork to solve their next frontier of problems.

Data Shama (@thedatashaman) 's Twitter Profile Photo

canadians don’t know you can work in sf, take the ferry, enjoy good weather and get paid more than $90k a year without paying 52% of your income in taxes to write code

canadians don’t know you can work in sf, take the ferry, enjoy good weather and get paid more than $90k a year without paying 52% of your income in taxes to write code