Projects

Product, data, and AI work across real systems.

Featured

Flagship

AI-Powered Fitness & Nutrition Platform

Cross-platform fitness ecosystem combining workout tracking, nutrition logging, analytics, and AI-powered personalization. Built as a monorepo spanning web, mobile, shared packages, and Firebase backend services.

ReactReact NativeExpoTypeScriptFirebaseOpenAI API

• Workout tracking, sets/reps/weight logging, and training volume history
• Nutrition tracking, food search, quick-add functionality, and calorie/macronutrient analysis
• AI meal descriptions and AI meal suggestions based on user goals and context
• Shared data between web and mobile with Firebase auth, Firestore, and cloud-backed logic

Architecture

Web App

↓

Shared App Layer

↓

Firebase / Cloud Functions

↓

AI + Nutrition Services

↑

Mobile App

GitHub →

FitAdapt

In Development

Transparent fitness intelligence engine that turns user profiles and longitudinal data into adaptive energy estimates, trend analysis, and personalized recommendations.

Python 3.12pytestRuffuvREST API

• Adaptive TDEE estimation based on observed trends and user context
• Longitudinal analysis of weight, intake, steps, sleep, and training activity
• Deterministic recommendation logic designed for transparency and reproducibility
• API-focused Python engineering with testing and clear calculation boundaries

View GitHub →

Experience

MapleJet — AI Engineer

Developed an internal AI-powered support assistant using Botpress, structured knowledge bases, and workflow automation to route product questions and support technical inquiries grounded in the appropriate documentation.

BotpressAI WorkflowsKnowledge BasesJavaScript

Additional projects

SpaceX Landing Prediction

End-to-end data science project exploring launch data, identifying factors associated with successful first-stage landings, and visualizing trends through interactive dashboards and predictive modeling.

PythonPandasSQLPlotly

Automobile Sales Analytics Dashboard

Interactive dashboard for comparing automobile sales across recession and non-recession periods, with filtering and trend exploration for historical data analysis.

PythonDashPlotlyPandas