WorkoutDB — Personal Fitness Tracking Application

Stack: C# (ASP.NET Core Minimal API), Azure SQL Database, Azure App Service, Chart.js, T-SQL

Status: Deployed and in active use

Overview

A full-stack workout tracking application built from scratch with no prior C# experience, using AI-assisted development to accelerate learning. Planned and managed solo using Scrum methodology —> full Product Backlog, Sprint Backlog, and Definition of Ready/Done written before any code.

Problem

Workout tracking in Excel had hit its limits:

• Fixed columns per exercise (couldn't adapt when the routine changed)
• No way to track partial completion (fewer sets/reps than planned)
• Manual, error-prone data entry (scrolling a spreadsheet mid-workout)
• Growing null values with no clear structure (different exercises performed every day)

Data Model

Normalized relational schema in Azure SQL Database:

Table	Purpose
DailySummary	One row per day: sleep hours, sleep quality, abs completion, notes
Exercises	One row per exercise performed: links to a day, tracks actual vs. planned sets/reps/weight
ExerciseDefinitions	Reusable exercise templates: name, category, default sets/reps

Relationships: DailySummary (1) → (many) Exercises on DayID ; ExerciseDefinitions (1) → (many) Exercises on ExerciseDefID

Architecture

Browser (phone or desktop)
   ↓
ASP.NET Core Minimal API (C#)
   ↓
Azure SQL Database (WorkoutDB)
   ↓
Chart.js dashboards (rendered client-side from JSON API endpoints)

Hosted on Azure App Service, so it's accessible from any browser, which means no local network or Remote Desktop workaround needed.

Features

• Log actual vs. planned sets, reps, and weight per exercise
• Multi-exercise entry in a single form submission
• Duplicate exercise prevention
• Workout location tracking (dropdown)
• Daily summary: sleep hours, sleep quality, abs completion
• Two in-app dashboards:
  • Overview (/dashboard) — trends across all workouts
  • Per-exercise (/dashboard/exercise) — progression on a single lift

Design Decisions & Pivots

• Power BI → in-app dashboards: Started with Power BI, but its mobile app requires a work/school Microsoft account. Rebuilt the dashboards directly in the ASP.NET app with Chart.js instead — one less moving part, and works on any phone browser.
• Row-based exercise tracking: Avoided a fixed column per exercise so the schema stays flexible as the routine changes over time.
• Planning before code: Wrote a 21-story Product Backlog across epics, plus Sprint Backlog and Definition of Ready/Done, before writing any C#.

SQL Highlights

Development Approach

• Solo Scrum: Product Backlog → Sprint Backlog → sprint execution, tracked on a formula-driven Kanban board (Excel FILTER/VSTACK)
• Iterative, AI-assisted build — one feature/bug at a time
• Legacy Excel data cleaned via T-SQL (null backfill, name normalization, category corrections) before import

Repo Structure

WorkoutDB/
├── README.md
├── src/               # ASP.NET Core source (C#)
├── sql/               # Schema creation + key T-SQL queries
├── docs/              # ER diagram, architecture notes
└── screenshots/       # Dashboard and app UI

Live App

https://1workoutinputapp20260702104629-fjg4hsc9hhd7cdbg.centralus-01.azurewebsites.net/

What I Learned

• Building a relational schema from scratch and normalizing it as requirements changed
• ASP.NET Core minimal API design and deployment to Azure App Service
• Azure networking/firewall configuration (rules live at the server level, not the database)
• Applying Scrum discipline to a solo project