research_desk
Jev demo that reads live yfinance profiles and headlines into ranked, grounded and routed trade ideas, showing every request.
Overview
Research Desk is a TypeSafe Jev demonstration for fast analysis of news and tickers. It loads live company profiles, prices and headlines from yfinance, projects them into state, and runs a five-stage pipeline in which Jev reads them into ranked, grounded and routed trade ideas. Every number in the UI is a typed answer from one request, and a Requests tab shows the exact state and questions behind each.
Questions and the thresholds code decides on live in one file, and a FastAPI server holds the API key and streams stage events over SSE. An autonomous demo tour is included. The README states plainly that nothing here is investment advice and that thresholds are starting guesses, not fitted to real outcomes.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Score | How relevant or material is this headline to the ticker?(inferred) | — |
| Choice | Which route should this trade idea take?(inferred) | — |
How it works
- Load profiles, prices and headlines from yfinance
- Ask Jev typed questions across five pipeline stages
- Apply code thresholds to rank, ground and route ideas
- Show results and the exact requests in the UI
Strengths
- Full transparency of state and questions per answer
- All thresholds in one file
Caveats
- Demo only; not investment advice
- Thresholds are unfitted starting guesses
- No licence stated
Getting started
cp .env.example .env (add TYPESAFE_API_KEY), then cd app && ./run.sh to serve on http://127.0.0.1:8765.At a glance
- Type
- App
- Status
- prototype
- Language
- Python
- Runtime
- FastAPI web app
- Has code
- yes
Use cases and integrations
How curated lists describe it
TypeSafe Jev demonstration for new analyzation — experimenting with Jev for fast analysis of news and tickers.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev decide?
It answers typed questions that rank, ground and route trade ideas from live headlines and profiles.
Can I see what Jev was asked?
Yes, the Requests tab shows the exact state and questions behind every number.
Is it financial advice?
No, the README says so explicitly.
Related projects
Chinese open-source research and trading workbench whose JEV monitor watches futures contracts and judges entries and exits.
ProjectbetaTypeScriptSelf-hosted trading analysis platform pairing Claude briefings with Jev probabilities for signals like thesis-breaking headlines.
ServicebetaStock terminal that gathers live evidence via Valyu and asks Jev for the probability a stock trades higher in 30 days.
ProjectprototypeTypeScript- QuantDinger★ 12,754
Self-hosted open-source AI trading OS with research, backtesting and paper/live execution, advertising Jev-powered pre-trade decisions.
LibraryproductionPython Watches X for updates relevant to your YAML profile, has Jev classify and score each post, and alerts in Slack.
ProjectbetaTypeScriptWorking paper placing a text judge beside a live trading system as a read-only signal, comparing Jev and Claude as blind headline judges.
Paperresearch