Building a framework for AI called BAML. (YC W23). @BoundaryML

Joined August 2011
383 Photos and videos
Pinned Tweet
We are hiring Rust devs to work on building out a programming language (BAML) with crazy new, useful features for making AI pipelines. Our entire team is senior engineers. We are a team of 5. We work hard and in person in Seattle. We accept new grads. The interview is not leetcode, but a 1-week work trial. Send me a DM on X or LinkedIn or email aaron @ boundaryml dot com if interested.
2
5
32
5,546
Aaron Villalpando retweeted
BAML is now an official language on @github with syntax highlighting. We made the sheep instantly recognizable 🐑🚀 Congrats @boundaryML
6
9
92
8,799
AI slop phrases from Claude: - ceremony - load-bearing - not x, not y, not z
2
119
claude fable's writing sounds like AI Slop turned to 11
1
85
all software tools must be made for agents
1
43
if your cli doesn't default to agent-mode already it's ngmi
71
Aaron Villalpando retweeted
At @boundaryML We're building building a full compiler, VM and async runtime. Kinda like TypeScript V8 Node.
May 28
i wonder about people who are claudemaxxing/codexmaxxing 18 hours a day, like what are these folks building exactly?
3
3
17
3,416
Could someone try rewriting TS in Go with Claude workflows please
96
We are launching a programming language built for agents soon called BAML that has been in the making for 1 years. You can follow @boundaryML We are a small team of 6 developing it with care, gathering feedback from humans and agents. Fully Open Source. If you are interested DM me for early alpha access.
Introducing Zero The programming language for agents. I wanted a systems language that was faster, smaller, and easier for agents to use and repair. Explicit capabilities. JSON diagnostics. Typed safe fixes. Made for agents on day zero.
2
2
13
1,554
Aaron Villalpando retweeted
Imagine if v8 was tied to the visualization, and you could watch your code execute. or click on teh graph and jump to that line of code. Or don't imagine and try this early beta in BAML.
Ask Claude to document and describe the main flows in your app and output in a single page html json data file. Incredibly useful for humans and the JSON file is very useful for explaining the flow to the LLM when working on new features/bugfixes.
2
2
15
3,039
Design is the bottleneck now, so we have a Kanban exclusively for more large-scale design docs (in our case it's language proposals).
1
123
imagine having to learn this
140
👀👀👀 what if we had Zig's IO (0.15), but simpler syntax
2
151
added full type signatures to llm functions to our dashboard. works with any input/output type. imagine querying dashboard.get<MyFunction>(last=5) with a full a typesafe client that works across diff versions of your function. Agents are gonna love this.
93
sorry we cant release BAML 1.0 yet, it's too powerful for the general public
1
1
10
691
is it just me or might tanstack be doing way too much
59
6
918
115,439
Aaron Villalpando retweeted
Announcing the BAML Bounty... For all power-users of BAML, we're giving away free BAML merch! (t-shirts, stickers, hoodies 🔥🧯). Share what you built with BAML with #baml → Fill out tally.so/r/PdErze → Free merch! Hurry! Supplies are limited to the first 50 posts.
5
6
885
my fav kind of chart
1
1
6
510
one is called autoresearch, the other one is called GEPA
It’s quite unfortunate that GEPA Optimize Anything didn’t get enough traction, while very, very similar ideas promoted by Karpathy’s autoresearch Lütke’s pi-autoresearch - got so much traction, despite being less general
2
375
your senior distinguished principal software engineers know as much about vibe coding as the SDE2 vibe coding on side projects. This stuff is all new. It's ~1 year old
5
374
code is turning into a live, self-evolving organism
2
370