Curriculum Vitae

Manoj Dalui

Quant Analyst & Python Developer  ·  Algorithmic Trading Systems

Get in Touch
Email
info@quantfinco.com
LinkedIn
manoj-dalui
Location
India
Certification
QuantInsti EPAT – Dec 2025
Professional Summary

Seasoned software engineer with 17+ years of enterprise IT experience who transitioned into quantitative finance through self-directed learning and formal training at QuantInsti (EPAT, Dec 2025). Designed and built a full quantitative trading ecosystem from scratch — comprising 7 independent backtested strategies, a production daily portfolio pipeline with macro regime classification, a 14-module loss-position decision engine, and a full-stack web platform for interactive backtesting and risk analytics. All systems built from first principles to deeply understand the underlying mathematics, covering NSE, BSE, and global markets.

Key Projects
QuantFinco – Quant Trading Web Platform Full-Stack
Flask · Python · Plotly · Bootstrap 5
  • Built a complete Flask web application with public portfolio pages and a protected Quant Lab dashboard secured via Flask-Login authentication.
  • Implemented 6 interactive dashboard pages: Run Backtest, Strategy Library, Market Explorer, Strategy Optimizer, Risk Metrics, and Portfolio Simulator.
  • Market Explorer computes SMA 20/50/200, RSI(14), MACD(12,26,9), rolling volatility, and 12 fundamental stat cards — all on-the-fly from yfinance with a data quality % score.
  • Strategy Optimizer performs parameter grid search across all 7 strategies using itertools.product, ranking results by Sharpe, return, and drawdown.
  • Risk Metrics page computes VaR(95%), CVaR, Sharpe, Sortino, Alpha, Beta, Calmar, Skewness, Kurtosis, rolling 60-day Beta, monthly returns heatmap, and return distribution histogram with normal-curve overlay.
  • Portfolio Simulator accepts comma-separated tickers, fetches data from yfinance with Angel One fallback, simulates weighted equity vs Nifty 50, and reports Sharpe, max drawdown, and annualised volatility.
  • Integrated SMTP email via GoDaddy Workspace (info@quantfinco.com) for contact form delivery.
  • Deployed via Gunicorn production server; frontend charts via Plotly.js and HTML5 Canvas.
7-Strategy Backtesting Framework Quant Research
Python · pandas · yfinance · Angel One SmartAPI
  • Designed a modular 4-layer architecture (indicators → strategy → backtester → optimizer) implemented consistently across all 7 strategies.
  • Strategies implemented: MA Triple Crossover (SMA 20/40/80), RSI Mean Reversion (Wilder EWM, SL/TP logic), Bollinger Band mean reversion, High-Low Breakout (N-day channel), Momentum (threshold + time exit), Turtle Trading (ATR-scaled SL/TP, long & short), Buy-Sell-Next-Day (consecutive down-day reversal).
  • Universal DataLoader with dual-source fallback: yfinance primary → Angel One SmartAPI for Indian equities, with 98% trading-day coverage validation and corporate-action adjustment.
  • State-machine signal generation with zero lookahead bias; metrics include log-return CAGR, Sharpe, max drawdown, win rate, return multiplier, and trade log with entry/exit prices and P&L %.
  • Parameter grid search optimizer ranks all combinations; results exposed via both CLI and QuantFinco web API.
Geopolitical Macro Investment Strategy Engine Production Pipeline
Angel One SmartAPI · FRED · NewsAPI · Jinja2
  • Built a 7-step daily pipeline (runs at 18:30 IST on trading days): data ingestion → feature engineering → macro regime classification → signal scoring → portfolio construction → historical backtest → HTML report & alerts.
  • Data ingestion aggregates from 4 sources: Angel One SmartAPI (150+ NSE/BSE stocks), yfinance (fundamentals, global ETFs), FRED API (US 10Y yield, Fed Funds rate), NewsAPI (geopolitical headlines + TextBlob sentiment).
  • Macro regime classifier identifies 4 states — GEOPOLITICAL_STRESS, OIL_SHOCK, RISK_OFF, NEUTRAL_BULL — with dynamic sector-weight multipliers across 18 sectors (defence, IT, pharma, energy, metals, etc.).
  • 4-component composite signal scoring: Momentum 30% (6M return rank) + Fundamental 25% (P/E, P/B, D/E, revenue growth) + Macro Alignment 25% (regime sector weight) + Technical 20% (RSI, MA position, Bollinger).
  • Portfolio builder applies position-mode sizing (conservative 2–5% → high conviction 20–40%), 25% sector cap, ATR-based stop-loss/take-profit, and cash buffer management.
  • Generates daily HTML reports (Jinja2) and email/alert notifications; persists state to JSON for multi-day continuity.
Loss-Making Stock Decision Engine Decision Support
Python · Transformers · GARCH · Rich TUI
  • Developed a 14-module advisory system to determine whether to HOLD, EXIT GRADUALLY, or EXIT NOW a loss-making position.
  • Modules cover: technical analysis (trend, support/resistance), fundamental valuation, GARCH(1,1) volatility forecast, peer relative-strength, news sentiment (FinBERT/TextBlob), corporate governance, promoter & institutional holdings, seasonal patterns, and portfolio-level risk concentration.
  • Each module produces a 0–1 confidence score; aggregate weighted score drives the final recommendation with plain-English reasoning.
  • Interactive Rich TUI (terminal UI); reads actual holdings from my_holdings.json automatically — user inputs only the ticker.
Quantitative Skills
Strategy Design: MA Crossover, RSI Mean Reversion, Bollinger Bands, High-Low Breakout, Momentum, Turtle Trading, Buy-Sell-Next-Day
Signal Engineering: 4-component composite scoring (Momentum, Fundamental, Macro Alignment, Technical); percentile-rank normalisation
Macro Analysis: Regime classification (4 states), sector-weight multipliers, macro feature extraction from FRED & RBI DBIE
Statistical Methods: GARCH(1,1), VaR/CVaR (95%), Sharpe, Sortino, Calmar, Alpha, Beta, Skewness, Kurtosis, rolling 60-day Beta
Backtesting: Log-return vectorised backtest, zero lookahead bias, transaction cost & slippage modelling, walk-forward optimisation
Risk Management: Position sizing (4 modes), sector capping (25%), ATR-scaled SL/TP, max drawdown monitoring, portfolio stress testing
NLP / Sentiment: FinBERT transformer sentiment, TextBlob polarity, NewsAPI geopolitical headline aggregation
Market Data: Angel One SmartAPI (primary), yfinance (fundamentals + fallback), FRED API (US macro), NSE/BSE/Global coverage
Technical Skills
Python: pandas, NumPy, SciPy, arch (GARCH), statsmodels, yfinance, smartapi-python, fredapi, newsapi-python, textblob, transformers, torch, pyotp, pyyaml, python-dotenv, rich
Web & API: Flask, Flask-Login, Jinja2, Bootstrap 5, HTML5, CSS3, JavaScript, Plotly.js, HTML5 Canvas, REST APIs, smtplib, Gunicorn
Data & Storage: yfinance API, Angel One SmartAPI, FRED API, NewsAPI, JSON state persistence, YAML config, CSV/pandas pipelines
Infrastructure: Git, Gunicorn (production), Windows Task Scheduler (daily pipeline), .env secrets management, structured JSON logging
Education & Certification
Executive Programme in Algorithmic Trading (EPAT) Completed Dec 2025
QuantInsti Quantitative Learning Pvt. Ltd.
Quantitative strategies, statistical arbitrage, risk management, ML in finance, options pricing, live market implementation using Python.
Bachelor of Engineering – Computer Science / Information Technology 2007
Foundation in algorithms, data structures, software engineering principles — applied directly to system design of all trading platforms above.
By The Numbers
7
Trading Strategies Backtested
4
Macro Regimes Classified
14
Decision Modules (Loss Engine)
6
Dashboard Pages (Web Platform)
150+
Stocks in Investment Universe
30+
Technical Indicators Computed
17+
Years of IT Experience
4
Live Data Sources Integrated
All code was developed independently using personal knowledge and public resources only. Personal project · Not financial advice.