Syed Ayan

Computer Science Student · Data Science Specialization · Full Stack & ML Enthusiast

_{Undergraduate studying Computer Science (Data Science) • Exploring Machine Learning, Data Analytics, Scalable Systems & Open Source}

"learning. failing. improving. repeating. every line of code tells a story."

I am an undergraduate student pursuing Computer Science with a specialization in Data Science (CS-DS). I am passionate about understanding data patterns, building predictive models, and bridging the gap between data science, machine learning, and scalable software systems.

As a student, I believe the best way to master computer science is by building real, end-to-end applications—applying theoretical mathematics, statistics, and algorithms directly to practical problems.

• Data Science & Machine Learning: Exploratory Data Analysis (EDA), statistical modeling, feature engineering, and predictive pipelines with Scikit-Learn and Python.
• Data Engineering & Analytics: High-performance columnar querying with DuckDB, dbt SQL transformations, and Parquet data pipelines.
• Full-Stack Development: Building robust web backends with FastAPI and Node.js, coupled with clean, interactive user interfaces in React.
• Applied Algorithms & Systems: Implementing core Data Structures & Algorithms (DSA), distributed IoT architectures, and exploring Web3 technologies.

A showcase of hands-on data science pipelines, analytical tools, and full-stack software built throughout my studies:

Tech: DuckDB 1.3.2 · dbt 1.12.5 · Python 3.12 · Scikit-Learn · Streamlit · Parquet · GNU Make · uv

• Sub-12ms Query Latency: Columnar analytics engine built on DuckDB, executing aggregations over millions of rows in milliseconds.
• End-to-End Orchestration: Automated single-command lifecycle (make all): raw ETL → dbt compilation → ML training → batch scoring → warehouse updates.
• Data Quality & Testing: Verified with 43 dbt model test assertions + 39 pytest unit tests passing with 100% test coverage.
• Streaming Telemetry Simulator: Real-time carrier signal simulator (Jio, Airtel, Vi, BSNL) joined directly with customer risk scoring.
• Interactive Dashboard: Responsive Streamlit interface with animated count-up KPI tiles, factor coefficient waterfall breakdowns, and live risk queues.

Tech: FastAPI · React 18 · PostgreSQL · Vite · Docker · WebSockets · Python · Machine Learning

• Fraud Intelligence: Integrated machine learning fraud detection model alongside Anti-Money Laundering (AML) graph analytics for detecting suspicious transaction rings.
• DSA Execution Engines: Built around 7 native Data Structures & Algorithms (DSA) execution engines for transaction scheduling, graph traversals, and ledger reconciliation.
• Architecture & Security: Digital banking interface with multi-factor authentication (JWT + MFA) and 5-role Role-Based Access Control (RBAC).
• Real-Time Telemetry: Live bidirectional WebSocket stream powering live transaction approvals and audit trails.
• Containerization: Multi-stage Docker setup for reproducible local development and deployment parity.

Tech: Python 3.10+ · TypeScript · FastAPI · React / Next.js · SQLite · HMAC-SHA256 · Capacitor (Android)

• Edge Risk Engine: Rule-based clinical triage scoring executed locally on edge nodes and identically on the backend to prevent score drift.
• Store-and-Forward Resilience: Self-healing routing topology with automatic rerouting to alternate relays during simulated network partitions.
• Cryptographic Security: Gateway-level HMAC-SHA256 frame verification ensuring payload integrity.
• Testing & Verification: Validated through 249 comprehensive tests and a 40-session recorded experimental evaluation run.
• Cross-Platform Client: Desktop app, installable Progressive Web App (PWA), and Android build via Capacitor.

Tech: HTML5 Canvas 2D · Vanilla JavaScript (ES Modules) · Vite · Web Audio API · CSS3 Glassmorphism

• 60 FPS Canvas Physics: Custom 2D HTML5 Canvas engine rendering real-time ship movement, trade routes, and dynamic weather physics.
• Economic Simulator: Dynamic supply-and-demand market simulation modeling fluctuating prices for commodities across Mediterranean ports.
• Procedural Sound Engine: Zero-dependency audio synthesizer using Web Audio API for environmental soundscapes and sound effects.

Tech: PHP · MySQL · JavaScript · HTML5 / CSS3

• Privacy-Preserving Surveys: Decoupled student identity submission ensuring 100% candid academic and faculty feedback.
• Analytics & Dashboards: Aggregate rating scoring, sentiment trend visualization, and department-level evaluation metrics.

Tech: Python · JavaScript · HTML5 / CSS3 · Makefile

• Automated Space Allocation: Algorithmic bay assignment maximizing facility space utilization.
• Real-Time Telemetry: Occupancy tracking and live slot availability visualization.

I'm always keen to connect with fellow students, developers, researchers, and mentors. Whether you want to talk about Data Science, Machine Learning, Open Source, or discuss collaboration opportunities and internships:

     

Copyright (c) 2026 Syed Ayan. All rights reserved.

🎓 Notice Regarding Academic Submissions & Plagiarism

This repository and profile represent original research, implementation, and portfolio work created by Syed Ayan.

• Portfolio & Evaluation Use: This profile and associated documentation are publicly available for technical review, peer exploration, and prospective evaluation.
• Prohibition of Plagiarized Academic Submission: Unauthorized duplication, reproduction, or submission of original project architectures, text, or implementations as academic coursework or institutional degree submissions is strictly prohibited.
• Third-Party Rights: All third-party marks, badges, and open-source dependencies remain subject to their respective licenses and terms.