Scoring 723 ads against 30 simulated buyers with Jev (X)

Still from Scoring ads against 4,000 simulated buyers with Jev (X) — Still from x.com

Post on X

Views: 131,332
Likes: 1,153
Reposts: 78

Read from X on 24 Sep 2026. Counts change daily.

Reported by the author

Decisions: 21,69021,690 stop or scroll decisionsx.com
Total cost: 22 cents21,690 stop or scroll decisions. 22 cents.x.com

Matthew Berman had Jev scroll through 723 ads as 30 simulated buyer personalities. For each pairing of persona and ad, Jev decides whether that buyer would stop on the ad or keep scrolling.

That works out to 21,690 stop-or-scroll decisions, which Berman reports cost 22 cents in total. He pitches it as a replacement for the focus group. The post says it will be available in StealAds and through an MCP server, a standard way for AI agents to call outside tools. It follows an earlier post where Berman had Jev break down 724 live ads from 37 brands by hook, format, offer, CTA, awareness stage and landing-page mismatch in 40 seconds, for 9 cents of tokens.

Open on X

View the post on X

PrimitivesScore

Use casesComposite scoring

jev-trader

Makes one AI trade decision every Monad block on the Kuru MON-USDC market using Jev. The public deployment runs in dry-run mode with a mock model.

jarrodwattsOpen

2.2k starsLoop, mock model p50 100 msDecision window ~300 ms/block

ProjectConfidence-gated actions+1

Introducing System One Models and Jev (Hacker News launch thread)

The 1,978-point, 513-comment launch thread. The top comment calls the speed comparison misleading, since Jev only returns structured output, and disputes the cannot-hallucinate claim because a valid type can still hold a wrong value. Others note ordinary LLMs can already be forced into structured output.

albelfioOpen

DiscussionIntent and model routing+3

DSPy: programming, not prompting, language models

Declares typed input/output Signatures for LLM modules and optimizes the underlying prompts and weights against a metric. Its Signature abstraction is the closest widely-used open equivalent of Jev's typed-question interface. Star count is GitHub's rounded display figure.

stanfordnlp38k starsOpen

AlternativeStructured extraction+3

newsjack

Open source skills that read a morning's news feed with Jev and tell a list of brands which stories to act on. In a demo run it read 384 headlines and matched them to 15 brands in 24.9 seconds for $0.19, against 4 of 384 for $0.77 on Claude Opus 5 over the same feed.

Elvis (@elvissun)Open

851 starsCost $0.19vs Opus/headline ~390x cheaper

ProjectComposite scoring

decider: one-pass typed decisions with calibrated probabilities (Qwen3.5-2B)

Open reproduction of the System One model class, fine-tuned from Qwen3.5-2B-Base, returning a probability distribution per typed question in a single forward pass. Serves TypeSafe's POST /v1/systemone wire format so TypeSafe SDKs work unchanged by repointing TYPESAFE_BASE_URL.

Mapika348 starsOpen

AlternativeIntent and model routing+3

pg-jev

PostgreSQL extension that lets you filter, rank and classify rows with plain-language conditions, such as WHERE jev(tickets, 'the customer is angry'). Adds jev_prob, jev_choice and jev_score functions that compose with joins, GROUP BY and ORDER BY, with no index or embeddings. Published on PGXN.

realZachiOpen

329 stars2k rows, cold ~3.5s, $0.0122k rows, cached ~50ms

ProjectSupport inbox triage+1

Menu