💻 Final Year Computer Science & Engineering Student @ Manipal Institute of Technology
🤖 Pursuing Minor Specialization in Computational Intelligence
🚀 Passionate about Artificial Intelligence & Machine Learning

✨ I love building intelligent systems that combine AI + Machine Learning to solve real-world problems.
🔍 Currently exploring LLMs, Computer Vision, and Blockchain Security, while contributing to open-source projects.
🌱 On a mission to learn, build, and innovate every day.

Amazon Jan 2026- Jun 2026

• Worked with large-scale internal Amazon systems and operational tools to validate, audit, and process high-volume transactional data, ensuring accuracy, compliance, and system integrity across workflows.
• Performed root-cause analysis on data mismatches and system-generated exceptions by analyzing logs, reports, and rule-based outputs, contributing to process optimization and defect reduction.
• Utilized automation-driven workflows, dashboards, and rule engines to monitor operational KPIs, detect anomalies, and support data-backed decision making for business stakeholders.

Xcitium / Comodo Cybersecurity Solutions Nov 2025 – Jan 2026

• Built data-driven trading intelligence models leveraging market microstructure gamma-exposure analytics.
• Improved trade-outcome prediction accuracy by 30% through optimized ML pipelines and semantically rich composite features.
• Automated an oscillator-based signal-detection framework, achieving 5× faster feature extraction.
• Collaborated with research leads to integrate structural and statistical indicators into production-grade models.

JPMorgan Chase Jun 2024 – Jul 2024

• 🌐 HTTP Request Tracking: Tracked and logged HTTP requests and responses (URLs, headers, payloads) into a secure MySQL database using Java & Spring Boot.
• 📊 Data Visualization Dashboard: Created a web dashboard with Chart.js, displaying metrics such as number of requests per domain and total data size transferred.
• 🔐 Security & Authentication: Implemented user authentication and data encryption with Spring Security to ensure privacy and controlled dashboard access.

A quantitative trading oriented time series analysis framework designed to systematically extract, identify, expand, and statistically validate recurring market and trade setup patterns from financial price series, returns, indicators, and derived trading signals.

• 🛠️ Built a leakage-safe sliding-window framework to extract and engineer trading features from financial time series
• 🧠 Discovered recurring market patterns using matrix profiles and expanded setups via similarity and DTW
• 🌳 Statistically validated and visualized trade structures to assess stability and robustness

Tech Stack: Feature-Engineering Feature-Extraction Pattern Mining TSFresh

Leakage-safe feature engineering, decision tree based clustering, interpretable rule extraction, and rigorous multi-stage validation using renko dataset. The workflow systematically reduces raw features into high-quality, production-ready trading patterns, emphasizing explainability, robustness, and out of sample reliability.

• 🛠️ Built an end-to-end quantitative ML pipeline to transform raw financial data into interpretable and production-ready trading patterns using Renko datasets
• 🧠 Engineered leakage-safe features with multi-stage reduction (190 → 50 features) ensuring robustness, reduced overfitting, and strong out-of-sample performance
• 🌳 Trained a decision tree–based clustering model generating 195 pattern clusters and extracted 42 high-confidence, rule-based trading signals

Tech Stack: Feature-Engineering Feature-Extraction Pattern Mining TSFresh

An AI-driven healthcare workflow built with n8n that automates insurance eligibility checks and pre-authorization using LLMs, APIs, and event-based orchestration.

• 🤖 Automated insurance pre-authorization using n8n and AI, cutting manual effort by 65%.
• 🔄 Orchestrated real-time eligibility checks with AI-generated clinical justifications for faster submissions.
• 🗄️ Enabled approval tracking, proactive alerts, and operational analytics using PostgreSQL.

Tech Stack: Postgres Json PostgreSQL automation sql n8n-workflow openai-api supabase supabase-auth

This project is a simple, self-contained Retrieval-Augmented Generation (RAG) service. Given a starting URL, it crawls a website, indexes its content, and answers questions strictly based on the information it has collected, providing citations for its answers.

• 🛠️ Engineered an End-to-End RAG Pipeline using FastAPI and Scrapy to ingest and index web content, utilizing ChromaDB for vector storage and Sentence Transformers for semantic search.
• 🌐 Integrated a local Llama 3 model via Ollama, applying strict prompt engineering guardrails
• ☁️ Optimized Data Ingestion & Retrieval.

Tech Stack: python Json PostgreSQL Sqlite Webhooks json-server rag generative-ai chromadb rag-pipeline llama3-8b

A Machine Learning Model forecasting air quality

• 🏗️ Designed and deployed a machine learning-powered web application for real-time air quality forecasting, addressing rising concerns over public health and climate change..
• 📊 Leveraged historical AQI, PM2.5, and PM10 datasets with advanced feature engineering to train deep learning models, enabling accurate future pollution predictions.
• ⚡ Integrated TensorFlow.js for seamless browser-based real-time inference, ensuring fast, accessible, and scalable deployment.

Tech Stack: TensorFlow.js Python JupyterNotebook Keras Pandas Deep Neural Networks

TensorFlow-Powered Machine Learning Model for Classifying Pedigree Charts into Autosomal Dominant, Autosomal Recessive, X-Linked Dominant, X-Linked Recessive Y-Linked Dominant, & Y-Linked Recessive Inheritance Pattern.

• 🛠️ Built a custom pedigree chart dataset modeling autosomal dominant/recessive, X-linked dominant/recessive, and Ylinked dominant/recessive inheritance patterns.
• 📈 Trained and optimized a TensorFlow classification model with preprocessing, normalization, and one-hot encoding ,using the Adam optimizer for faster convergence.
• 🎯 Applied hyperparameter tuning and cross-validation, achieving 94% accuracy in predicting inheritance patterns on unseen data.

Tech Stack: TensorFlow.js Python JupyterNotebook Keras Pandas Convolutional Neural Networks

A Flutter app for IoT based Arduino nano non-invasive device for comprehensive blood glucose monitoring.

• 📱 Developed an end-to-end industrial IoT health monitoring app using Flutter/Dart with Firebase for real-time sync.
• </> Applied Concepts of Software Engineering Like SDLC Prototype Model in development, achieving 95% alignment with objectives and improving development efficiency by 40% through iterative prototype refinement.
• ⚡ Conducted performance optimization tests, improving system efficiency by 30%.

Tech Stack: Flutter Dart Software Engineering Google Firebase Firestone RestAPI

Flutter app for table reservation database management

• 📱 Developed a cross-platform mobile application using Flutter and Dart, integrating Firebase for real time data synchronization and authentication, which enhanced user engagement by 30%.
• 🎨 Designed intuitive UI/UX for Table reservation based mobile app, resulting in 25 % increase in user satisfaction and a 20 % reduction in user churn.
• 🔗 Implemented REST APIs for seamless communication between mobile app and Backend services, reducing data retrieval time by 50 %.

Tech Stack: Flutter Dart JSON Google Firebase Firestone RestAPI

• Bachelor of Technology, Computer Science and Engineering
• Minor Specialization in Computational Intelligence

Manipal Institute of Technology • Sep 2022 – July 2026

• Relevant Coursework:
  • Data Structures & Algorithms
  • Object Oriented Programming
  • Operating Systems
  • Computer Networks
  • Soft Computing Paradigms
  • Artificial Intelligence
  • Machine Learning
  • Computer Vision
  • Parallel Computing
  • Blockchain Technology

• Pollution Predictor : A machine Learning Model For Forecasting Air Quality (Minor Project Publication )
• Medisure Type 2 Diabetes Predictor (Undergoing)

• IN23/2404 Application No. 202441091887 Device and method for monitoring blood parameters of a user (Early Publication available)