We Rewrote JSONata with AI in a Day, Saved $500K/Year

Simon Willison's Blog / 3/27/2026

💬 OpinionDeveloper Stack & InfrastructureSignals & Early TrendsTools & Practical Usage

Key Points

  • Reco’s team “vibe-ported” JSONata by creating a new custom Go implementation of the JSON expression language, aiming for behavior parity with the existing reference implementation.
  • They leveraged JSONata’s existing test suite to get the first working Go version running in about 7 hours, reportedly costing around $400 in token spend.
  • To validate correctness, they used a shadow deployment that ran the old and new implementations in parallel for a week.
  • The case study highlights how reuse of mature tests plus AI-assisted development can accelerate re-implementations and reduce ongoing engineering cost (claimed $500K/year savings).
Sponsored by: WorkOS — Ready to sell to Enterprise clients? Build and ship securely with WorkOS.

27th March 2026 - Link Blog

We Rewrote JSONata with AI in a Day, Saved $500K/Year. Bit of a hyperbolic framing but this looks like another case study of vibe porting, this time spinning up a new custom Go implementation of the JSONata JSON expression language - similar in focus to jq, and heavily associated with the Node-RED platform.

As with other vibe-porting projects the key enabling factor was JSONata's existing test suite, which helped build the first working Go version in 7 hours and $400 of token spend.

The Reco team then used a shadow deployment for a week to run the new and old versions in parallel to confirm the new implementation exactly matched the behavior of the old one.

Posted 27th March 2026 at 12:35 am

This is a link post by Simon Willison, posted on 27th March 2026.

go 51 json 146 ai 1934 generative-ai 1715 llms 1681 agentic-engineering 34 vibe-porting 7

Monthly briefing

Sponsor me for $10/month and get a curated email digest of the month's most important LLM developments.

Pay me to send you less!

Sponsor & subscribe