COMMUNITY OBSERVEDMetric claims: AUTHOR REPORTEDSOURCE AUDITED🤖 Jev Agent

Personal Feed Filtering with Jev

Turn a noisy social feed into a shortlist of posts worth reading.

Overview & result

Ackerman demonstrates a personal feed filter using Jev to evaluate relevance, value, format and unsupported claims. The author reports narrowing 1,000 posts to 20. The reported 97 ms per post and 1.2 cents for a morning are author claims, not measurements reproduced by this site.

From the original post

98% of my feed is junk. Now I don’t even see it

I built a filter that reads my feed for me. It runs on Jev - a model that doesn’t generate text; it just makes binary decisions, in milliseconds and for pennies.

For every post, it evaluates four questions: Is it relevant to my niche? Does it provide real value? What format is it (breakdown, news, ad, meme)? And does it make loud claims with zero proof?

Out of 1,000 posts, only 20 remained.

97 milliseconds per post.

1.2 cents for the entire morning.

The most frustrating takeaway, nearly half of my feed was ads and memes, not the creators I originally followed for substance.

An hour of mindless scrolling turned into five lines over breakfast.

— Ackerman

How Jev fits in the loop

  1. The author’s filter evaluates each feed post.
  2. Jev answers the author’s questions about relevance, value, format and evidence behind claims.
  3. The filter keeps a smaller reading list based on those judgments.

How to reproduce

  1. Watch the original filter demonstration and read the four evaluation questions.
  2. The post does not link an implementation repository. Before building a similar filter, define what is relevant to your own reading goal and manually review excluded posts.

Why this build matters

Shows how explicit reading preferences can reduce the time spent scanning an unfiltered feed.

Limitations

  • Reported counts, speed and cost have not been independently reproduced.
  • The author’s preferences are not a universal definition of useful content; relevant posts may be excluded.

Patterns