devwithjevNews Feed Classification Pipeline
Ns rebuilt an AI knowledge graph as a news feed, using Jev as the decision model in the classification pipeline. The model cannot generate text, and the new stack uses Postgres and two small services.
View on X cost~10x cheaper
ns@nicky_sap๐
i killed my ai knowledge graph and rebuilt it as a news feed. the whole classification pipeline now runs on a decision model that can't generate text and it's ~10x cheaper. the old stack: neo4j + redis + a claude pipeline doing 8k-token extractions on every story. ~$40/week, and i shut it down because it was too expensive. the new stack: postgres, two tiny services, and jev (@typesafeai's system o
The previous stack used Neo4j, Redis, and a Claude pipeline that performed 8k-token extractions on every story. Ns said it cost about $40 per week and was shut down as too expensive.
Also filed under Content & growth
- Transcript-Based Video Highlight Editing
jevclip is a video editing pipeline where Jev judges transcript segments to select highlights and remove filler.
- SEO AIO GEO Dashboard
Karo says the SEO AIO GEO Dashboard was built with Jev.
- Viral Hook Telemetry for Social Media
Hookmeter-Jev is a Chrome extension and JEV System 1 for millisecond-level viral hook telemetry and social media co-piloting.
- Jev Website SEO Auditor
It audits any website for SEO from one homepage URL, with Jev judging the results. It produces PDF, XLSX, and Markdown reports.