Harvard's Gary King Team Launches Certify: Agent-Automated Replication Certification for Journals
RexDouglass · x · 2026-11-05
Gary King, Yiqing Xu and Leo Yang launched Certify, software that adds a replication certificate to a dataset—proof that rerunning the submitted code produces the paper's published results.
Why it matters
- Authors avoid months of publication delays and years of reader questions.
- Journals skip full reanalysis: data editors can verify in minutes that figures and tables match reported numbers.
- Readers can confirm replicability in about a minute; certified datasets get cited roughly 2x more.
How it works
- Authors submit replication packages to Harvard Dataverse as usual; certification is automated by agents and runs once on the author's own machine, sealing a certificate on the dataset page.
- Nothing is public until the author clicks publish; the folder is never modified.
- Currently in beta on demo.dataverse.org, with feedback being solicited.
Related event: Gary King team introduces cryptographic replication certificates for papers(3 posts)→
More from coding & agent
- Claude spontaneously finds a serious OpenBLAS bug while just explaining code to a user — Aizkmusic · 2026-11-05
- Next-gen task management: agents should book human attention via MCP — lxfater · 2026-11-05
- GraphRAG in real projects: 3 practical patterns from Daan Di Scala's webinar — Al_Grigor · 2026-11-05
- Orderain ships an MCP server that lets Claude run your ecommerce store on autopilot — Slight_Guest3459 · 2026-11-05
- OpenAI 'Presence' page surfaces, pushing trusted AI agents into customer channels — juberti · 2026-11-05
- RL teaches agents to use their own notes; training survives 23M tokens of memory edits — simran_s_arora · 2026-11-05