AI multi-dimensional research

19 AI analysts,
parallel equity research

Five rule-based analysts plus thirteen investor-master AI personas independently analyze fundamentals, technicals, valuation, market sentiment, and master views. A LangGraph DAG runs them in parallel, then aggregates the signals into a quantifiable consolidated report. Supports A-share and US equities.

Try it online

600519 Kweichow Moutai · Consolidated Report

72 / 100
Fundamentals ROE 30.2% · Margin 52.1% Bullish
Technicals EMA golden cross · ADX 38 Bullish
Valuation P/E 32.5 · DCF premium 8% Neutral
Sentiment Institutional buying · Positive news Bullish
Buffett — Bullish Munger — Bullish Graham — Neutral Peter Lynch — Bullish Burry — Neutral Damodaran — Bullish

Core capabilities

19 analysts reach independent conclusions; multi-dimensional aggregation removes single-view bias

📊

Five-dimensional independent analysis

Fundamentals (ROE / margin / growth), technicals (EMA / RSI / ADX / Bollinger), valuation (DCF / EV / EBITDA), sentiment (insider trading + news), and master views — each scored independently.

🧠

13 investor-master AIs

Buffett on the moat, Graham on margin of safety, Peter Lynch on PEG, Burry on contrarian bets, Taleb on tail risk — each master applies their own investment philosophy to reach an independent verdict.

LangGraph parallel orchestration

19 analysts run in parallel through a LangGraph fan-out / fan-in DAG; after risk review and signal aggregation, they produce a consolidated report that doesn't rely on any single view.

🏧

A-share + US equities with real data

A-share uses baostock for real data (free, no API key needed); US equities support multiple data sources. Flip a provider via an env var without touching agent code.

📈

Async analysis + historical lookback

A Celery task queue runs analyses asynchronously while the frontend polls without blocking. All results are persisted in PostgreSQL and can be revisited by ticker, date, or score.

🔒

Risk review and disclaimers

A risk-review agent inspects extreme signals independently before aggregation. Every report is labeled as AI-generated and carries a disclaimer: this is not investment advice.

13 investor-master AIs

Each master applies their own philosophy; different viewpoints combine into a fuller picture

Warren Buffett

Moat + owner-earnings DCF

Charlie Munger

Multi-model thinking + quality first

Ben Graham

Margin of safety + Net-Net value

Peter Lynch

PEG + know what you own

Phil Fisher

Long-term growth + management quality

Michael Burry

Deep value + contrarian investing

Cathie Wood

Disruptive innovation + growth

Druckenmiller

Macro trends + momentum

Bill Ackman

Activist investing + FCF

Nassim Taleb

Antifragility + tail risk

Damodaran

DCF + CAPM intrinsic value

Mohnish Pabrai

Downside protection + FCF yield

Jhunjhunwala

Margin of safety + emerging markets

Workflow

Enter a ticker → 19 parallel analyses → aggregated output

1

Enter a ticker

A-share code or US ticker

2

Data collection

baostock / yfinance fetch price and financial data

3

19 parallel analyses

LangGraph DAG fan-out; each agent reaches its own verdict

4

Risk review

Checks extreme signals and data consistency

5

Signal aggregation

Vote counting + weighted score + consolidated report

Let AI do the equity research homework

Enter a ticker; 19 AI analysts run in parallel and produce a multi-dimensional consolidated report in minutes. Free to try, A-share and US equities supported.

Visit AI Equity Research

Note: Reports are generated automatically by AI models for informational and educational purposes only. This is not investment advice. Investing carries risk; please proceed with care.

Explore more products