Quantitative developer & fullstack engineer.
Trading systems, data pipelines, software that moves fast.
I'm Mohammed Sahal, an Information Technology student at Cochin University of Science and Technology. I'm passionate about quantitative finance, algorithmic trading, and building data-driven systems that operate at scale.
My work spans from low-latency C++ trading engines to full-stack web applications. I specialize in Python, C++, and machine learning frameworks, with a focus on creating efficient solutions for financial markets.
Develop and backtest quantitative trading strategies for global markets. Design data pipelines, optimize strategies using performance metrics and market insights.
Designed and implemented the company website and admin panel. Worked with Oracle APEX, Odoo ERP, Firebase, and MERN stack.
Contributed to open-source projects, enhancing functionality and fixing bugs.
Managed and developed the organization's website — 25% increase in user retention. Implemented chatrooms.
Multi-interface download manager aggregating Ethernet, Wi-Fi, and 5G into a single high-performance stream. Up to 43× faster than a browser. 9-phase execution pipeline, adaptive scheduling, SHA-256 integrity verification, Electron GUI.
Analyzes real-time political sentiment from Reddit, Twitter, and financial news to predict market movements. Custom NER model with fuzzy matching, VADER/TextBlob scoring, LSTM forecasting, PMI correlation models.
Detects illegal insider trading using Isolation Forest and Autoencoders. Integrates real-time market data and SEC filings. Granger causality and ADF tests for statistical validation.
Low-latency C++ trading engine with a custom order book for real-time price matching. Statistical arbitrage strategies and backtesting on historical market data.
High-precision Formula 1 prediction system using Monte Carlo simulations and historical race data. Probabilistic modeling of racing scenarios with optimized simulation parameters.
Real-time credit card fraud detection using Random Forest and Gradient Boosting. Adaptive learning mechanisms and risk scoring to minimize false positives over time.
Jira-Trello clone with drag-and-drop functionality and subtask management. Built with React and CSS.
Platform for sharing portfolios among peers and recruiters. User auth, portfolio sharing, real-time updates.
Automated healthcare data collection for Asha workers. C++ desktop app with Next.js, TailwindCSS, Prisma, and PostgreSQL. Chart visualization and real-time analytics.