#142 Why LLMs Need Their Own Programming Language: From Assembly to AI with Vaibhav Gupta // Co-founder @ BAML
Topics discussed
Introduction and Vaibhav Gupta's background in AR and YC
Pivoting from Slack competitor to building BAML
Early coding experiences and AI's impact on velocity
Solving hard problems vs. boilerplate code with AI
Why BAML is a new paradigm for LLM compute
BAML features: Type safety, evals, and observability
Shipping code at machine speed and versioning
Comparison to Google's Protobuf and developer tooling
Live coding demo: Generating BAML with Claude
Visualizing codebases and shifting trust in CI/CD
The future of engineering: Agents and abstraction
Democratization of building and the future of SaaS
Composable systems and the rise of non-terminal users
Personal advice on retirement, happiness, and Bitcoin
Conclusion and AlphaList podcast outro
Listen ad-free on Castria