Programming Languages as Core Technology for AI
summary
Basis describes a concrete stack: Cairo causal PPL built on Pyro, an effectful Python library for algebraic effects, and program synthesis integrated into the MARA agent. Code represents world models, interventions become a do handler that rewrites sampling, and counterfactuals run as a multi-world counterfactual handler that threads shared randomness. Implementation targets staged emission to PyTorch (and potentially DAX) so most effects are compiled away for performance.
AI-written summary · about
related talks
34mProgramming Languages for AI Programing Agents
32mProgramming language evolution in the AI era
24mSoftware 3.0 and the AI Engineering Landscape
26mAl and Software: The Good, The Bad, The Scary
14mFull-Stack AI Engineering from AI-Native Coding to Knowledge, Models, Infrastructure & OSS
48mDyalog and AI
28mSoftware 3.0 and The Emerging AI Developer Landscape
30mToward Programming Languages for Reasoning -- Humans, Symbolic Systems, and AI ...