GitHub Next's PDD Kit uses formal methods to prove AI-generated web app code is correct
0xkarasy · x · 2026-11-05
GitHub Next, in collaboration with Microsoft Research, released PDD Kit (Proof-Driven Development Kit), a research prototype addressing a sharpening problem: coding agents now write code faster than anyone can review line by line.
- The shift: Formal methods have traditionally been reserved for safety-critical systems like operating systems, hypervisors, and databases. PDD Kit extends them to full-stack web apps — per the Stack Overflow survey, 44.4% of developers work in full-stack, back-end, or front-end roles.
- How it works: The kit converts what the developer meant into a Verus specification, then mathematically proves the AI-generated code meets that spec for every input, supplemented by tests where a proof isn't possible.
- Human/agent division of labor: Humans own the judgment behind what the software should do; exhaustive correctness checking falls to mathematical proof.
Team: Alex Lavaee, Matthai Philipose, Maggie Appleton, Krzysztof Cieślak. Full blog post is up on githubnext.com.
More from coding & agent
- Ghosthand gives AI agents a real iPhone to beat captchas and bot detection — ethanniser · 2026-11-05
- OpenAI ships ChatGPT Images 2.6 with GIFs, reference images, and 50% faster API models — kagigz · 2026-11-05
- Biomedical benchmark: prewritten Cypher MCP tools hit 98% vs 85% for Text2Cypher — JeremyCMorgan · 2026-11-05
- Musk shows off Grok calling a dentist to check insurance coverage — elonmusk · 2026-11-05
- Treat AI bots as employees, not assistants: an onboarding playbook — elonmusk · 2026-11-05
- Codex Voice cuts tool latency by streaming context to the harness before delegation — SIGKITTEN · 2026-11-05