MPC Quant AI Engine — Phase 1

Institutional-grade AI trading intelligence.

Upload your OHLCV dataset. Our Chronos-2 forecasting engine and XGBoost ML model analyse every bar and return a precise BUY / SELL / NO TRADE signal with full indicator breakdown.

Start Analysis How it works

MPC Quant Portal

AI ENGINE v1.0

LIVE

Sample Output — NIFTY 50 / 5M

◈ NO TRADE

Current

23,775.10

Forecast

23,773.52

Return

-0.0100%

RSI 48.2 ADX 22.5 ATR 87.4 EMA200 +3.86%

Confidence 1%

Chronos-2 + XGBoost Fusion Run on your data →

208K+

Training Rows

2015–26

NIFTY Data Range

150.6%

Backtest Return

AI Features

Step 1

Upload Your Dataset

Accepts OHLCV CSV files — NIFTY 50, Stocks, Forex, Commodities

Drop your CSV file here

or click to browse your files

Required CSV Format

date	open	high	low	close	volume
09-01-2025 09:15	23,500.00	23,542.80	23,480.10	23,530.45	1,24,500
09-01-2025 09:20	23,530.45	23,568.20	23,510.00	23,551.70	98,200

Select Instrument

NIFTY 50

● Active

BANKNIFTY

Phase 3

Stocks

Phase 3

Crypto

Phase 3

Initialising MPC AI Engine...

Reading and cleaning OHLCV data

Calculating RSI, ATR, ADX, EMA, MACD

Running Chronos-2 price forecast

XGBoost return prediction

Generating signal decision

AI Analysis Complete

Signal Report

Generated at —

AI Signal Decision

◈ NO TRADE

Instrument

NIFTY 50

5-Minute

Current Price

—

Chronos Forecast

—

Combined Return

—

Chronos Return

—

Time-Series Forecast

XGBoost Return

—

ML Regression Prediction

Confidence Score

confidence

BUY threshold > 0.25%

SELL threshold < -0.25%

Trend Indicators

EMA 20 —

EMA 50 —

EMA 200 —

EMA200 Distance —

Momentum

RSI (14) —

MACD —

MACD Signal —

ADX (14) —

Data Summary

ATR (14) —

Last Close —

Data Rows —

Date Range —

FastAPI Response /signal

{}

Disclaimer: MPC Quant AI Engine is a research and analysis tool. Signals are for informational purposes only and do not constitute investment or financial advice. Past backtest performance does not guarantee future results. Please consult a SEBI-registered investment advisor before making any trading decisions.

MPC AI Architecture

How It Works

Data Ingestion

OHLCV CSV parsed and invalid candles removed.

Indicator Engine

RSI, ATR, ADX, EMA20/50/200, MACD — all 9 features computed.

Chronos-2

Last 512 bars fed to Chronos-2. Median from 21 quantiles extracted.

XGBoost ML

Trained on 166K+ rows. Predicts future return from all indicator features.

Signal Decision

Combined return averaged. BUY >0.25%, SELL <-0.25%, else NO TRADE.

Development Roadmap

Platform Evolution

Phase 1

Core Engine

✓ Data Pipeline
✓ Indicator Engine
✓ Chronos-2 Integration
✓ XGBoost Model
✓ FastAPI Backend
✓ Quant Portal UI

COMPLETE

Phase 2

CSV Engine

→ Auto column detection
→ Live CSV processing
→ Per-upload AI analysis
→ Interactive price chart

Phase 3

Multi-Asset

○ BANKNIFTY, FINNIFTY
○ Stocks & Commodities
○ Forex & Crypto
○ Auto instrument detection

PLANNED

Phase 4/5

Pro Platform

○ TradingView charts
○ Equity curves & Sharpe
○ User accounts
○ API subscriptions

FUTURE