stanford-mast/a1

The agent-to-code compiler that optimizes AI for speed 🏎️ and safety 🏁

6 best points · 2 comments · 13 Hacker News posts · 61 GitHub stars · View on GitHub

Hacker News discussions

Show HN: Agent-to-code JIT compiler for Z3-theorem-proving agents (6 points, 0 comments, by calebhwin, 2025-11-13)
Show HN: What if MCP agents were JIT compiled to code? (6 points, 0 comments, by ardmiller, 2025-11-14)
A1: Agents-to-Code JIT Compiler (5 points, 0 comments, by calebhwin, 2025-11-13)
Show HN: A1 – compiler for AI agents into maximally deterministic code (3 points, 1 comments, by calebhwin, 2025-12-08)
A1: Agent-to-Code JIT Compiler (2 points, 0 comments, by calebhwin, 2025-11-12)
JIT Compiling AI Agents to Code (2 points, 0 comments, by calebhwin, 2025-11-18)
Show HN: A1 – cost-optimizing JIT for CRUD AI agent translation to code (2 points, 0 comments, by calebhwin, 2026-01-02)
Show HN: a1 - determinism-maxing JIT compiler for AI agents (1 points, 0 comments, by calebhwin, 2025-11-10)
Show HN: A1 – An optimizing JIT compiler for AI agents (1 points, 0 comments, by calebhwin, 2025-11-11)
Show HN: Optimizing JIT Compiler for Code Mode MCP (1 points, 0 comments, by ardmiller, 2025-11-21)
Show HN: A1 – Local Sandbox and JIT Compiler for AI Agents (1 points, 1 comments, by calebhwin, 2025-11-27)
Show HN: We're Building an AOT/JIT Compiler for Program-of-Thought Prompting (1 points, 0 comments, by calebhwin, 2025-12-03)
An Optimizing JIT for LLM Tool-Use to Code (1 points, 0 comments, by calebhwin, 2026-01-09)