Making a programming language - BAML. @boundaryml, 🦄 aitw podcast: piped.video/@boundaryml prev YC, google, msft, deshaw, and other things

Seattle, WA
Slop is code you dont read. And better models means more slop. the solution? write sloppy tools to build stable systems. but for the most committed people, go build new foundational systems. Rebuild github, rebuild tmux, rebuild database, and yes rebuild programming languages. Big thanks to @swyx and the whole community for having us out at @aiDotEngineer
6
11
78
14,639
AI will never ever write code in assembly. Lets me tell you why. I spent most of my career writing assembly, and the hardest part of job was never to write things in ASM, it was deciding what to write in ASM. Writing things in ASM has a real tax. You lock in architectures, memory layouts, and making your code fast, means making it more rigid. Think about our supply chains during covid. Shipping stops and suddenly everything breaks. Speed is never free. AI might write some code in ASM, but only if the value you're getting is worth the benefit. New folks can benefit from SSE/AVX instructions, but the majority of code in the world needs to remain flexible and updatable. At 1M Tokens / second, would you rather generate a for loop in ASM? or an entire app in Python? Time will always be money and compilers and higher level languages are going to win. Hopefully BAML :)
Replying to @vaibcode
It looks very interesting, great work! But on the other hand, I doubt that in a year we'll still be writing and reading code ourselves. Why do we need another programming language if AI will probably be writing code directly in assembly?
3
12
1,399
Build composable layers so you can incrementally rewrite it! Have found this to be very true in our codebase as well.
We rewrote an internal thing from Zig to JavaScript and it got faster and more reliable. The obvious conclusion is that language doesn’t matter. Rewriting does. You should be rewriting everything all the time.
6
581
Vaibhav Gupta retweeted
bullish ‘bout baml honestly
Sounds like we need a new language 🤔 Boundaryml.com
1
9
1,212
Sounds like we need a new language 🤔 Boundaryml.com
JavaScript was never designed for the server. If that doesn't change soon, JavaScript on the server will be replaced by Rust. The fix needs to happen in the engine. Sound types. Ahead-of-time compilation. Threads with shared objects.
25
12
356
63,732
Vaibhav Gupta retweeted
schema-constrainted generation has/had downsides. @boundaryML's take from last year: boundaryml.com/blog/structur… @vaibcode, founder + team have worked on just this problem, 99% OSS, just added Veb support. @HanchungLee thoughts?
1
1
2
223
Vaibhav Gupta retweeted
Had great fun talking with @vaibcode and Kai from BAML about how performance engineering is total kryptonite for freestyle coding agents! piped.video/watch?v=WHLXRZi8…
1
9
516
i'm just always seeing "context automatically compacting" longer context windows when?
4
5
980
I think we’re looking for this github.com/BoundaryML/feelin…
putting an early bet that Jev’s biggest use will be some flavor of context window protection for agents by filtering tool results agents will perceive a cleaner world through the lens of Jev
3
2
39
3,879
A new challenger has appeared: Javascript
Java is the most disgusting stupid ugly horrible language to ever be invented
1
3
859
Vaibhav Gupta retweeted
If you didn't know, this is the ultimate win. Jev is a fantastic model for a number of use-cases. BAML is agentic programming superpowers Together they will be MAGIC!
we're also fans! jev as an AI function coming to BAML in tonights release! github.com/BoundaryML/baml/p…
1
1
290
we're also fans! jev as an AI function coming to BAML in tonights release! github.com/BoundaryML/baml/p…
Replying to @remilouf
so did langchain! (not an insult, I'm a fan of both and BAML/instructor/guidance/all of it) the problem with building on TOP of LLMs is you inherit the flaws in their optimization (RLHF/RLVR) and still pay the cost of text generation we gave up a lot (text generation) to push the intelligence per $ and second pareto frontiers by a ton
5
3
44
3,747
damn got me.
4
13
995
code is always the source of truth i can't wait until harnesses are post-trained to do this more
1
8
489
this is your daily reminder to read something. read the specs read the plans read the code read the prs model drift is certain.
2
25
690
find yourself an @antoniosarosi
Hey @vaibcode, you should promote @AntonioSarosi to CMO. First @RogerKernel, now Rubén. A lot of people in my Spanish TL are talking about @boundaryML due to Antonio's marketing.
1
4
1,111
you've heard of IsRecord, but have you seen CookedRecord?
2
1
10
723
no more queuing messages in side chats apparently!
1
2
599
its amazing when you make your code so fast you uncover a race condition
2
1
11
918
plan mode came and went so fast
2
12
754
agent infra is just an append only log
agent infra is just all global singletons
3
17
2,510