Joined June 2009
195 Photos and videos
An interesting sub-thread on the value (or lack thereof) of language integration
Replying to @yminsky @sebngriego
"Language integrated" is a drum I've been beating on for a while [e.g. with refinement types youtube.com/watch?v=F2tYCxb3…] but in the age of LLMs I wonder if *really* matters, if the AIs are going to also be generating the proofs?
1
1
28
7,775
Yaron (Ron) Minsky retweeted
Hell yes
Our goals here are ambitious! Our hope is to make formal methods as pervasively useful of a tool for building software as sophisticated type systems are for us today. blog.janestreet.com/formal-m…
1
1
35
7,128
Yaron (Ron) Minsky retweeted
Replying to @yminsky
Even a mixed approach is still ambitious, but it will likely be worth the effort. There are a lot of unexpected nuances to formal verification. It's a very fun field to work in!
1
6
3,186
Our goals here are ambitious! Our hope is to make formal methods as pervasively useful of a tool for building software as sophisticated type systems are for us today. blog.janestreet.com/formal-m…
16
36
245
21,620
Yaron (Ron) Minsky retweeted
Replying to @yminsky
Also, it's a nice place to work, can't leave that out!
1
3
1,427
One natural question if you're a formal methods person: why is Jane Street the right place to do this kind of work? I think we have two things going for us: deep control of the language we're using, and a community of programmers who are ready for this. blog.janestreet.com/formal-m…
7
9
131
7,770
Of course, we can do some pretty cool things with type systems! Here's a nice talk from Dolan on that: youtube.com/watch?v=W5li5LBY… Indeed, our experience with agents and types is part of what makes us excited to see what we can do with yet more powerful methods.
14
93
9,058
Yaron (Ron) Minsky retweeted
Replying to @yminsky
This is super-exciting. It's a great moment for formal methods.
1
23
4,152
I've been telling people for 25 years that Jane Street is not interested in formal methods. No more! And we're actively hiring to form a new formal methods team!
29
72
1,067
120,016
Yaron (Ron) Minsky retweeted
1. In my experience, the art is in constraining agents just enough and no more.
1
1
3
1,284
Yaron (Ron) Minsky retweeted
Had an interview with Jane Street in 2026... ...the other kind of interview😂
Much fun talking with @natefoster about networks, BGP, programming languages, formal methods, and what it's like being a visiting researcher at Jane Street.
1
3
83
8,790
Much fun talking with @natefoster about networks, BGP, programming languages, formal methods, and what it's like being a visiting researcher at Jane Street.
5
7
110
14,788
The actual video is here: youtube.com/watch?v=CHHZwari…

1
1
20
3,738
Yaron (Ron) Minsky retweeted
Replying to @yminsky
Nice! And purely out of self interest, this is a nice collection of examples that I can try `bombadil terminal` on. Trying to get the opam stuff to run on NixOS now...
1
1
9
2,214
Yaron (Ron) Minsky retweeted
BadUserBench, where all the prompts are contradictory or bad and the model has to figure out what to actually do
10
6
106
18,190
Yaron (Ron) Minsky retweeted
Expect tests are amazing for AI code review.
1
8
2,796