Open to opportunities

Amey somvanshi

@ameysomvanshi

Message

Engineering Physics student focused on quantitative finance, ML, and scalable AI-driven products.

India

Message

What I'm looking for

I seek roles building production ML/AI systems or quantitative models where I can deliver scalable, low-latency features, collaborate cross-functionally, and grow in ML engineering and finance.

I am a B.Tech (Engineering Physics) student at IIT Roorkee with hands-on experience building ML systems, full-stack AI features, and quantitative finance models. I have shipped production-ready AI tutoring features, optimized latency to sub-2s, and developed high-performing models for credit risk, sentiment-driven trading, and image classification.

My work blends rigorous data science, software engineering, and product delivery — from sensor-integrated mobile apps for agriculture to quantum-classical models and robust trading strategies. I thrive in collaborative roles where I can drive measurable impact through model tuning, prompt engineering, and scalable infrastructure.

Experience

Work history, roles, and key accomplishments

Research Intern

Technopedia

Jun 2025 - Aug 2025 (2 months)

Built a news-headline sentiment pipeline (228K+ headlines) combined with technical indicators to forecast stocks and designed a strategy turning $10K into substantial multi-year returns with high Sharpe ratios.

Web Scraping NLP Sentiment Analysis Time Series Python Data Engineering

Data Science Intern

Luner

Jun 2025 - Jul 2025 (1 month)

Led product, infrastructure, and ML systems for an AI tutoring platform; shipped full-stack features (RAG doubt solving, diagram generation, adaptive learning) and optimized latency to sub-2s at scale.

Python Prompt Engineering Model Fine Tuning RAG Full Stack ML Infrastructure

Data Science Intern

Finance Club, IIT Roorkee

May 2025 - Jun 2025 (1 month)

Preprocessed ~30K records and rebalanced classes with SMOTE; benchmarked models and achieved ~88% recall for defaulters using XGBoost for credit risk scoring.

data preprocessing SMOTE XGBoost Logistic Regression Random Forest Model Evaluation F2 Score

Undergraduate Researcher

IIT Roorkee

Feb 2025 - May 2025 (3 months)

Engineered a mobile app and sensor integration for plant health detection achieving 89% accuracy and real-time soil analysis under 2s, delivering personalized crop-care recommendations.

ESP32 Bluetooth Embedded Systems Mobile App Development Machine Learning Python

Researcher (Quantum ML)

IIT Roorkee

Jan 2025 - May 2025 (4 months)

Developed a quantum-classical hybrid model using TensorFlow Quantum for image classification, achieving 69.34% test hinge accuracy while reducing trainable parameters by ~99% versus classical CNNs.

Python Hybrid Adam Optimizer MNIST

Education

Degrees, certifications, and relevant coursework

Indian Institute of Technology Roorkee

Bachelor of Technology, Engineering Physics

Grade: 7.685

Activities and societies: Projects in Smart Plant Health, Credit Card Behavior Prediction, Stock Sentiment Analysis, Image Classification using Quantum Neural Network; roles include Placement Coordinator and Head of Operations (Bhawan App).

Pursuing B.Tech. in Engineering Physics (IV Year I Semester) with coursework and projects in quantitative finance, machine learning, and data science; cumulative CGPA 7.685 as of 2024.