back_office_ops · saas · workflow
Wix automates classification and conversion of 2,000+ Velo API code samples with GPT
A Velo syntax change rendered a huge number of Wix's 2,000+ API code samples outdated, and updating them manually would have required an enormous amount of tech writer work, halting the whole team's velocity.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Search source repos for samples
A script uses the GitHub CLI to search Wix's source repos for code sample files and saves them in a working repo with a mapping file.
Tools used
GPT-4GPT-3.5 turboGitHub CLITypeScript compiler
Outcome
By automating with GPT, Wix limited team involvement to 6 writers for one morning, the LLM produced no conversion errors, and reusable knowledge and code have since helped the team complete similar tasks much faster.
Results
Volumeover 2,000
Grounding & classification
Source type: technical build writeup
26 fields verified against source quotes.
code generationdocument classificationcode diff prknowledge basefailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedsource backedtools describedworkflow describedsoftwareemployee productivitytime savedtechnical build writeupback office opsextract classify routehuman review queue