About the role
We are building BAML -- a new programming language to build AI powered applications.
What does the code look like when 50% of the business decisions is decided by an AI agent / prompts? How do you test these AI components? What tools should we build?
Every computing paradigm, brought with a new language
- Desktop Computer → Holepunching and ASM were far too inaccessible → C
- Multiple OSs → Why write 3 variants of C → Java
- A computer on every desk → Lets make software more readable → Python
- Web browsers → Event driven / async native → JS
- AI / Graphics → We need to do a lot of math, and fast → Cuda
- Interactive Website → HTML + Logic is hard to do at scale → React
- Non-determinism → My software is isn’t predictable because AI → ?????
BAML is our answer, and it will let more developers than ever write AI pipelines without having to think about AI model reliability.
Why you should join us
Most startups die quietly. Most programming languages never leave a compiler talk.
But every once in a generation, someone builds the next C, Java, Python, JS, React or Cuda.
We’re trying to build that: a language for reasoning with AI.
We might fail spectacularly.
But if we don’t, it’ll be because a small group of curious, fearless builders decided to bet on beauty and correctness in an age of chaos.
If that excites you, then Boundary is your place.
But don’t take our word for it, try BAML first.
Responsibilities
- Since we’re building a compiler, you’ll be able to solve some of the hardest and most interesting problems — e.g. how to suspend/resume AI workflows, how to support calling BAML functions from any language, how to create a graph visualization from BAML code users or AI agents write, and exploring unique syntaxes.
