Raj

Projects

Apps, pipelines, models, and experiments.

24 projects
in-progress6 reports

Agricultural Data Analysis

Interactive USDA analytics dashboard with commodity price forecasting

Next.js 16React 19TypeScriptTailwind CSS 4RechartsDeck.gl
in-progress2 reports

APEX

Multi-signal ensemble algorithmic trading framework for US equities

Python 3.11+DuckDBpandasNumPyPyArrowLightGBM
in-progress2 reports

Ball View

Real-time cricket ball tracking overlay for live broadcasts

FastAPIOpenCVYOLOv8EasyOCRKalman FilterChrome Extension (Manifest V3)
live7 reports

CoverDrive Cricket

Analyst-grade IPL match intelligence for the serious cricket fan

Next.js 16TypeScriptTailwind CSS 4RechartsClerkFastAPI
live

CoverDrive XI Selector

Prescriptive analytics engine for optimal IPL playing XI selection

Python 3.11+PuLP (CBC solver)FastAPISQLAlchemy 2.0statsmodelsSciPy
in-progress4 reports

DataSkrive Cohort

Cohort audit and anomaly detection for a sports betting content platform

Python 3.14pandasNumPymatplotlibstatsmodelsscikit-learn
live2 reports

Horizon League Budget Dashboard

Athletic budget benchmarking platform for 11 NCAA Division I universities

Next.js 14TypeScriptTailwind CSS 3RechartsTanStack React TableAzure SQL (mssql/tedious)
in-progress3 reports

Intent Quotient Cricket

Quantifying batting intent in T20 cricket independent of outcome

Python 3pandasNumPymatplotlibseabornplotly
in-progress4 reports

Medicare Provider Cost Analysis

Predicting Medicare allowed amounts using provider-level features across 103M records

Python 3.xpandasNumPyPyArrowscikit-learnXGBoost
stable

Open Export

CLI tool to download and archive all your ChatGPT conversations

Python 3.10+PlaywrightClickRichhatchlingPyPI
alpha3 reports

Peruse AI

Local-first autonomous web agent for exploration, testing, and UX review

Python 3.10+PlaywrightasyncioOllamaLM StudioOpenAI-compatible APIs
completed

PlayerData

Athlete benchmark analytics framework for contextualizing collegiate wearable data

Python 3pandasNumPystatsmodelsStreamlitPlotly
in-progress2 reports

PocketLedger

Local-first personal finance tracker with OCR-powered statement parsing

FastAPISQLAlchemy 2.0AlembicPydantic 2.9+Tesseract (pytesseract)OpenCV
live

rvedire.com

Personal portfolio and project showcase site

Next.js 16React 19TypeScript 5Tailwind CSS 4Geist font familyMDX (next-mdx-remote 6.0)
published

PlayerData x IU Indianapolis Analytics

Data engineering and analytics pipeline for athlete tracking data, part of the IU Indianapolis–PlayerData partnership. Includes synthetic data generation, integrity checks, and agent-ready metric design.

PythonPandasData EngineeringSynthetic DataStatistical Analysis
published

GList — AI-Powered Grocery App

Android grocery list app with AI receipt scanning, Gemini-powered item categorization, and spending analytics. Built with Kotlin and Jetpack Compose.

KotlinJetpack ComposeGemini AICameraXML KitRoom
published

VacayStays — WhatsApp Property Bot

A WhatsApp Business bot for vacation property management with automated availability checking via iCal parsing, inquiry logging, and interactive guest responses.

PythonFastAPIWhatsApp Business APIiCalUvicornhttpx
published

RunIndex — Race Discovery Platform

A Next.js web application for discovering and browsing running races, with calendar views, geospatial search, and race detail pages.

Next.jsReactTypeScriptPostgreSQLPostGISSupabase
published

Project Antna — Mobile Survival Game

A mobile survival game built with Godot 4.5 featuring entity pooling, skill trees, experience progression, and wave-based enemy spawning.

Godot 4.5GDScriptGame DevelopmentMobile2D
published

NFL Big Data Bowl — Receiver Analytics

Advanced player-tracking analytics exploring wide receiver movement, separation dynamics, and post-throw performance using NFL tracking data.

PythonPandasScikit-LearnLSTMMatplotlibPlotly
published

PySpark Workshop — Batch ML to Real-Time Prediction

A hands-on 2-hour PySpark workshop covering distributed data processing, MLlib pipelines, MLflow tracking, and real-time streaming on Databricks.

PySparkMLlibMLflowStructured StreamingDatabricksDelta Lake
published

Video Game Recommender System

Content-based video game recommendation engine using IGDB data, cosine similarity, and a Streamlit interface for interactive exploration.

PythonScikit-LearnPandasStreamlitIGDB APICosine Similarity
published

Formula One Race Prediction

Machine learning pipeline to predict Formula One race winners using historical data from 1950–2020, achieving 87% prediction accuracy.

PythonPandasScikit-LearnMatplotlibJupyterErgast API
published

Plant Disease Classification

Deep learning image classification model using TensorFlow and transfer learning to identify plant diseases from leaf images.

PythonTensorFlowComputer VisionTransfer LearningResNet50Xception