RaghavSampath

M.S. Data Science @ Columbia University

"
"

A strong will is not something you're born with. It's something you decide to have.

ABOUT

About Me

Data Scientist passionate about machine learning, statistical modeling, and extracting actionable insights from complex datasets. Currently pursuing graduate studies at Columbia University with a focus on applied ML and data-driven decision making. I'm curently exploring the intersection of Agentic AI and finance and I aim to develop intelligent agents capable of navigating market dynamics with machine-level precision and human-like reasoning.

Education

🎓 MS Data Science @ Columbia University

Reading

📚 The Mamba Mentality: How I Play

Location

📍 New York, NY

EDUCATION

2021 - Jun 2025
Vellore, IN

Vellore Institute of Technology

Bachelor of Technology

Electronics & Communication Engineering

Relevant Coursework
Artificial Intelligence & Machine LearningProbability & StatisticsRandom Variables & ProcessesCalculusInformation TheoryComplex Variables & Linear AlgebraComputer Programming: Python
2025 - Dec 2026
New York City, NY

Columbia University

Master of Science

Data Science

Relevant Coursework
Exploratory Data Analysis & VisualizationAlgorithmsDatabase Management SystemsFinance & Structuring

WORK

Data Engineering Project Intern

Acceedo

Jan 2025 - Apr 2025
Coimbatore, IN

Built distributed ETL pipelines for real-time IoT systems, optimizing data ingestion and stream processing.

Migrated backend to Node.js, reducing system latency for sensor data processing
Designed NoSQL data models in MongoDB for high-availability analytical applications
Implemented RESTful API testing and automated validation for data governance

Machine Learning Engineer Intern

Quadrasystems.net

Jun 2024 - Aug 2024
Coimbatore, IN

Developed ML models for vulnerability scoring using NLP and deep learning techniques.

Built Vulnerability Score Predictor using Scikit-learn and PyTorch
Fine-tuned DistilBERT transformers, achieving 12% accuracy improvement
Deployed ML models on AWS EC2 with FastAPI serving real-time predictions

Project Intern

Steps Knowledge Services

Aug 2023
Coimbatore, IN

Worked on embedded systems and IoT projects integrating various hardware and cloud technologies.

Developed with ESP32 Dev Board for embedded systems applications
Built Amplify IoT project with Current Transformers and LoRa
Integrated ThingSpeak for real-time data monitoring and visualization

Academic Intern

National University of Singapore

Dec 2022
Singapore

Gained hands-on experience in data analytics, text mining, and deep learning methods.

Explored ANN, CNN, and RNN architectures for various AI applications
Applied text mining techniques for data analysis projects
Deepened understanding of neural network fundamentals and AI techniques

POSITIONS

Sep 2023 - May 2025Vellore, IN

Executive Member

VIT Student Council

Selected as 1 of 40 students out of 40,000 to serve on the Student Council for high-impact campus initiatives.

Coordinated logistics and judged finals at Yantra 2024 hackathon
Managed student grievances and stakeholder communications
Promoted from Member Secretary for leadership excellence
May 2023 - May 2025Vellore, IN

Core Committee Member

ACM-VIT

Led recruitment and event operations for the Research Domain of ACM's university chapter.

Designed puzzles for Cryptic Hunt '23 student engagement event
Reviewed team projects for HackStory at Yantra 2023
Coordinated technical events and student outreach activities
Mar 2022 - Mar 2024Vellore, IN

Core Committee Member

The Electronics Club

Supported student engagement and major event operations for the university's electronics community.

Hosted informational seminars for new member recruitment
Served as event anchor for SENSORED 6.0 flagship event
Managed A/V technical support and communications

SKILLS

Languages

PythonSQLJavaC++TypeScriptJavaScriptBashHTML/CSSRMATLAB

Frameworks & Tools

PyTorchTensorFlowLangChainLangGraphScikit-learnSparkHadoopPandasNumPyOpenCVSeabornMatplotlibStatsmodelsStreamlitFastAPIFlaskDockerGitAWSGCPHuggingfaceChromaDB

Methodologies

Machine LearningDeep LearningNLPTime-Series AnalysisStatistical ModelingA/B TestingLLMsReinforcement LearningComputer VisionMLOpsAnomaly DetectionFeature EngineeringDimensionality ReductionExperimental Design

RESEARCH

Patent

Real-time Foundry Environment Monitoring System

Jan 2, 2026

Integrated low-latency industrial IoT framework for real-time environmental sensing and power analytics in foundry environments. Features STM32-based microcontroller with interrupt-driven timing, WebSocket pipeline to MongoDB backend, and JWT authentication. Includes responsive dashboard with automated anomaly detection.

IoTSTM32WebSocketMongoDBReal-time Analytics
Conference Paper

Federated Learning for Predictive Maintenance: Model Comparisons and Privacy Advantages

Jan 2025
2024 International Conference on IoT Based Control Networks and Intelligent Systems (ICICNIS)

Presents a federated learning architecture for predictive maintenance, enabling multiple clients to train ML models while keeping data decentralized. Implements SMOTE for class imbalance, PCA for dimensionality reduction, and PSO for hyperparameter tuning.

Federated LearningPredictive MaintenanceSMOTEPCAPSO

PROJECTS

Morpheus

2026

Adversarial Strategy Simulation Engine.

OpenAI SDKHolonicPythonFastAPIGoogle Cloud RunTypeScript
View Code

Better Call Scount

2026

A multi-agent system using Google ADK that analyzes trending GitHub repos and tech news to hypothesize which tech stacks or industries are about to boom.

Google ADKRAGPythonFastAPIGoogle Cloud Run
View Code

OrderFlow

2026

An AI-powered order management platform. Employees can submit orders via voice or text, and the system uses Anthropic Claude to extract structured order data, detect anomalies, and generate customer quotes.

Next.jsFastAPIElevenLabs ScribePostgreSQLAnthrophic
View Code

Fuel Form

2026

Adaptive fitness and nutrition planning system that automates workout and dietary decisions using AI and machine learning. The goal of the application is to reduce the cognitive burden of planning workouts and meals by continuously adapting recommendations based on user behavior, recovery, and progress.

KotlinFlaskQualcomm AI Dev SDK
View Code

FX Trading

2025

Comprehensive FX trading system that combines traditional time-series techniques with modern machine learning methods for EUR/USD trading. The project includes both rule-based strategies (exponential smoothing with macroeconomic filters) and advanced ML approaches (ensemble models, LSTM, and regime-aware models).

PythonPyTorchXGBoostLightGBMStreamlitExponential Smoothing
View Code

Bends Before The Break

2025

Analysis of U.S. Treasury yield curves, inflation, unemployment, and their relationships with economic recessions. Uses data from FRED (Federal Reserve Economic Data) spanning 1954 to present.

Rggplot2Quatro
View Code

Kumori

2024

IoT Weather Monitoring System that collects environmental data using ESP32 sensors. Features data cleaning pipelines and predictive modeling for weather insights.

ESP32PythonMQTTInfluxDBStreamlit
View Code

P-Wagon

2024

Real-time vehicle detection system for law enforcement. Uses Raspberry Pi 4 to process surveillance footage with ML algorithms for license plate recognition and OCR.

PythonOpenCVTensorFlowRaspberry Pi
View Code

StudyEasy

2024

AI-powered study companion that generates flashcards, quizzes, and summaries from lecture notes and textbooks to enhance learning efficiency.

PythonFastAPIStreamlitHuggingfaceMistral
View Code

Amplify

2023

IoT-based energy monitoring system integrating Current Transformers with LoRa communication and ThingSpeak for real-time power consumption analytics.

ESP32LoRaThingSpeakC++
View Code

Tohsaka Rin

2022

Feature-rich Discord bot built with discord.py offering server management, entertainment commands, and utility functions for enhanced community engagement.

PythonDiscord.pyPostgreSQLDocker
View Code