Jev Daily

Treg ships a social data API priced on Jev filtering

Treg says his open source social API pulls 1,000 competitor ads for under $0.05 with Jev doing the filtering, and Diogo Almeida thinks that price is what finally makes running intelligence over big archived data worth doing.

Built with it

Treg's open source social API filters competitor ads with Jev

Treg released a Jev-powered social API that pulls social data through one endpoint, tracks mentions of you and competitors, and uses Jev to filter trending posts in a niche. He says it costs under $0.05 to pull 1,000 competitor ads, open source with 0% markup.

Diogo Almeida
@CompleteSkeptic
X
why don't we throw intelligence at the big data we already have? too expensive + too unreliable in map-reduces
Nov 7, 2026 · View on X

Read the full story

ILIAS ISM puts a two-step spend gate on his write-enabled MCP servers

ILIAS ISM upgraded all five of his products to remote MCP servers (tool protocol for agents) with write operations, not just GET wrappers that dump JSON into the chat. The backlink tool previews the exact quote and only charges the saved card when the agent passes `confirmPaymentSpend: true`.

ILIAS ISM
@illyism
X
they dump 20,000 tokens of JSON into your chat and then you still have to open 10 browser tabs to actually do the work
Nov 7, 2026 · View on X

Read the full story

Writing criteria

Moritz Kremb writes skill feedback into a doc the skill reads each run

Moritz Kremb says he now builds most of his agent skills with a feedback loop: one document the skill reads on every run. Whenever he has feedback on the output, it goes in that doc instead of a rewrite of the skill itself. He calls it the single biggest thing he has done to improve his skills.

Moritz Kremb
@moritzkremb
X
this way my skills self-improve over time
Nov 7, 2026 · View on X

Read the full story

Get the next one by email

Jev, read daily so you do not have to. The builds, the benchmarks, the criteria that worked and the cases where it lost, from the people shipping on TypeSafe AI's System One model.