
The programming language for AI
We are building BAML -- a programming language to build AI agents.
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?
BAML is our answer, and it will let more developers than ever write AI pipelines without having to think about AI model reliability.
We are looking for extremely talented engineers to make critical decisions on the language design, implementation, and the surrounding toolchain.
Be part of the first team to write a programming language built for the AI era. You will get to work on the BAML language itself as well as its powerful toolchain that’s used by thousands of developers every week.
You will work with an extremely talented group of senior+ engineers.
Send a message / email with Subject: “Why I’m awesome” with 3 bullet points with the best technical and/or personal achievements you are proud of. E.g. “I ran an ultramarathon in XYZ hours” or “I wrote a Rust crate used by millions of devs each month”. 1-5 sentences each.
Here is an example of one bullet point from one of our interns as reference:
“Compilers! Right after learning Rust, I began working on my most successful project so far: An ARM64 C compiler [1]. Generating code to and from SSA is much harder than I first thought, which led me to temporarily stop working on it. The C features it implements are pretty basic, and it made me realize that code generation is a hidden beast! Turns out the hard problem is not in typechecking or parsing C code. I've continued the compilers route to this date by working on my own inference engine prototypes [2] [3] and reading about Sea of Nodes [4] [5], among other prototypes that haven't seen the light, like attempting a Zig port of [1] while mimicking Zig's AST data structure layout [6], which proved interesting but not worth it due to having to constantly expand it.”
We created BAML -- a programming language to build AI agents.
C allowed us to stop thinking about Assembly Python freed us from manual memory management, letting us focus on building 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?
BAML is our answer, and it will let more developers than ever write AI pipelines without having to think as hard about AI model reliability.
We are looking for extremely talented engineers to make critical decisions on the language design, implementation, and the surrounding toolchain.