Quant Agent is an AI-powered stock research platform that continuously analyzes the top 50 US equities using a multi-stage pipeline. It combines technical indicators, fundamental data, news sentiment, social media signals, and macro context to produce actionable recommendations.
Each stock receives a composite score (0–100) built from multiple weighted dimensions:
Two AI analyses — short-term (1–4 weeks) and long-term (6–12 months) — each produce a BUY/HOLD/SELL call with a confidence level. Higher confidence amplifies the score impact.
RSI overbought/oversold levels, MACD crossovers, and price position relative to the 200-day moving average contribute directional signals.
Average of short-term and long-term AI sentiment scores (–100 to +100), reflecting overall market narrative and news tone.
P/E valuation relative to sector averages and revenue growth trends help distinguish overvalued from undervalued positions.
Sharpe ratio and risk-adjusted return metrics reward stocks with favorable risk/reward characteristics.
Bullish/bearish ratios from Reddit and social media discussions provide a real-time crowd-sentiment overlay.
Beyond individual scores, stocks are ranked through a pairwise tournament — a merge-sort algorithm where every comparison is an AI head-to-head evaluation.
The AI compares two stocks at a time across six dimensions, each with a specific weight:
Comparisons are cached daily, so repeated matchups resolve instantly. The tournament runs in scheduled phases throughout the trading day, progressively refining the ranking until all stocks are fully sorted.
Automated analysis runs five times per trading day (Mon–Fri), timed around key market sessions:
| Session | Time (approx. ET) | Purpose |
|---|---|---|
| Pre-Market | 8:30 AM | Set the stage before the opening bell |
| Intraday 1 | 11:30 AM | Early session momentum check |
| Intraday 2 | 1:30 PM | Midday reassessment |
| Intraday 3 | 3:30 PM | Pre-close positioning check |
| Closing Bell | 4:05 PM | End-of-day summary after market close |
Tournament ranking phases run staggered after each analysis session. A keep-warm timer ensures the platform stays responsive throughout market hours.
Market data and financials via Yahoo Finance. News aggregated from major financial RSS feeds. Social sentiment from Reddit communities. All AI analysis powered by Azure AI Foundry.