Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

open-jev

open-jev is a browser-focused TypeScript library for typed decisions: one piece of text (the state) plus any number of typed questions go in, and one forward pass returns a calibrated probability distribution per question. Nothing is generated, so an answer is always one of the options you provided.

Details

External ID
1378312690
Source
GITHUB
Company
—
Product
open-jev
Website domain
npmjs.com
Launched
Sept. 20, 2026
Cohort
—
Upvotes
33
Upvotes percentile
0.7449397899052012
Tags
—
Fetched at
Sept. 24, 2026, 5:02 p.m.
Updated at
Sept. 24, 2026, 5:02 p.m.

Enrichment

Theme
local AI inference and runtimes
Vertical
Horizontal
Function
Model & infra
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
browser typescript library for typed decision classification
Manually corrected
False

Could you build this?

Partial While wrapping browser-based ONNX runtimes in TypeScript is vibe-codeable, building a custom calibrated single-forward-pass classification head with proper calibration metrics requires ML research and training expertise.

What it would actually take: The architecture involves an in-browser WebGPU/ONNX Runtime engine paired with a quantized encoder model (like ModernBERT or DeBERTa) structured to output multi-target calibrated logits in a single inference pass. The hard parts are model training, temperature scaling/conformal prediction calibration across diverse question sets, and WebGPU memory budget optimization.

Competitors

Other products that read as similar to this one — 299 launches clear the similarity bar, closest 8 shown.

Attention rank: #125 of 300 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 323 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a model & infra tool for Fintech yet.