Hooking Jev into BigQuery: SQL rows scored as 0-1 probability columns via remote functions
rseroter · x · 2026-09-23
Developer Jeff Nelson connected typesafeai's Jev to BigQuery via a Cloud Run remote function: SQL queries call external code, and every row gets evaluated with Choice and Noul, turning nuanced judgment calls on unstructured data into 0–1 probability columns that can be filtered, aggregated, and joined in SQL.
The approach removes the need for manual labeling — model scores become queryable table fields. A blog post and GitHub repo were promised, and Google developer advocate rseroter amplified it as a great remote-functions use case.
More from coding & agent
- Grok Build 1.0.41 Ships with Grok 4.7, Adds Subagent Controls — elonmusk · 2026-09-23
- Every's GPT-6 Sol Hands-On: Near-Astra Writing, 50% Cheaper Than 5.6 Sol — every · 2026-09-23
- Agent Swarms Will Suffocate the Web, Making Bot Detection a Huge Market Gap — signulll · 2026-09-23
- Opus 5.5 is now live in Conductor — charlieholtz · 2026-09-23
- Reproducible agent evals: harbor makes configs, trajectories and logs shareable — seanwbren · 2026-09-23
- OpenAI launches GPT-6 Sol and Luna: near Fable 5 on DeepSWE at up to 96% lower cost — OpenAIDevs · 2026-09-23