Why two AI models instead of one chat model?

Architecture. Claude writes reasoning and briefings while Jev returns calibrated probabilities for in-pipeline signals such as whether a headline breaks a thesis. Without a Jev key those judgments degrade to a keyword classifier rather than failing.

How do you stop the AI from hallucinating prices?

It cannot invent numbers it did not fetch: portfolio chat, briefings, debates and scan explanations call live price reads, the scoring engine, news, options flow, insider signals, the discipline guardian, and the backtester. Free-form price or RSI recall from memory is designed out.

How does the AI bill stay around a dollar a month?

Haiku by default, one cached briefing per day, batched prompts, multi-hour caches per signal family, and a hard daily budget cap enforced in code that stops AI calls at your limit. The UI shows spend per day and the sample briefing day cost $0.14.

Can I trust the entry scores or are they curve-fit?

A 730-day backtester with walk-forward validation and bull/bear regime segmentation, plus forward logs for experimental signals. It already flagged one of its own thresholds as curve-fit, which is the point: thresholds must survive out-of-sample and both regimes.

How do I run TraderAI locally?

pip install -r requirements.txt, copy .env.example to .env, run the API with python -m uvicorn api:app --app-dir backend --reload --port 8000, and run the frontend with npm install && npm run dev. First launch opens the setup wizard.

What does the TypeSafe Jev component do?

It returns calibrated probabilities for thesis-breaking news, a live macro-narrative pulse, and big-move explanations. Without a key it degrades to a keyword classifier, so the Jev judgments are optional.

Is it investment advice, and does it trade?

No. It is analysis tooling for your own portfolio, run on your own machine, and it never executes trades; you confirm every action at your broker. It is MIT-licensed and keeps positions, rules, and history local.

CommunityJev Tools & CLIs

Traderai

Self-hosted trading intelligence platform pairing a Python scoring and risk engine with a Claude plus Jev analyst, where the analyst answers only through fetched data tools and the user confirms every action at their broker.

See highlights

Open github.com

Stars: 4
Language: Python
License: MIT
Published by: Community
Author: amanadhav
Last verified: Oct 2026

amanadhav/traderai

User repository on GitHub

View on GitHub

Self-hosted AI trading intelligence platform - scoring engine, two-model AI analyst (Claude + TypeSafe Jev), risk engine, discipline guardian, backtester, React dashboard

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Last spring I was holding a chip stock I really believed in. One Tuesday it jumped double digits overnight - earnings. I didn't even know they reported that night. The signals had been sitting in plain sight for days: unusual call buying, the supplier upstream had already beaten, prediction markets had the beat priced at 90%+. I just had no system watching any of it.

Quoted from the project's README. Read the full README.

Tagscommunity awesomejev catalog-import

Works forJev for Classification Jev for Moderation

Used inJev for Software & DevTools

Languages: Python

Highlights

180-point entry scoring over 14 factors plus a separate 150-point scorer for pre-profit small caps.
ATR-based stops and volatility-scaled sizing turn risk settings into enforced share counts.
Discipline guardian checks the live portfolio against your own caps, stops, cash floor and drawdown rules.
Onboarding wizard builds accounts, holdings via live ticker search, presets, themes and AI behavior.

Watch out

Analysis tooling only: it never executes trades and is not investment advice, with paper mode and local gitignored storage for positions and rules. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

jev-recall

Python memory filter that judges every stored memory against the current request with one calibrated yes/no per memory in a single Jev request, keeping everything above a threshold.

Jev Tools & CLIs#community#jevbest#catalog-import

Community

samdotmak/jev-recallTypeScript

Jev-Register-Tool

Pure-HTTP automation for the TypeSafe register, magic-link login, onboarding, and API-key creation flow, with pluggable temp-mail backends, concurrent batch runs, and key verification exports.

Jev Tools & CLIs#community#awesomejev#catalog-import

Community

2951461586/Jev-Register-ToolPython

dohnuts

Python toolkit plus 0.8B multimodal model that takes a message, document, or image and returns calibrated choice, yes/no, and score answers in one forward pass for local decision workloads.

Jev Tools & CLIs#community#awesomejev#catalog-import

Community

PsiACE/dohnutsPython

jevclip

Video tool where Jev scores subtitle segments and code assembles a short highlights cut plus a de-watered full cut with timestamped summaries.

Jev Tools & CLIs#community#jevbest#catalog-import

Community

cclank/jevclipPython

JEV-Paper-Radar

Forkable weekday radar that asks Jev one yes-or-no question per interest over each new arXiv announcement and publishes a page plus RSS feed.

Jev Tools & CLIs#community#jevbest#catalog-import

Community

Eliot5566/JEV-Paper-RadarPython

hermes-nerve

Async supervisory layer for Hermes Agent that locks Definitions of Done, proves mechanical criteria locally, and calls Jev only for unresolved semantic checks under a token budget.

Jev Tools & CLIs#community#jevbest#catalog-import

Community

keeltrace/hermes-nervePython

From the Jev guides

Original write-ups on what this entry is for.

Jev on the Gateways: Vercel, Workers AI, and OpenRouter

Oct 12, 2026

The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.

Jev Tools & CLIs#gateway#vercel#cloudflare

Read article

Exporting Decision Models: GGUF, ONNX, and Device Runtimes

Oct 10, 2026

Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.

Jev Tools & CLIs#open-models#local#onnx

Read article

Records at Rest: Documents, Inboxes, and Risk Queues

Sep 28, 2026

Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.

Jev Tools & CLIs#classification#email#security

Read article

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai