The latest articles on DEV Community by rahul kumar (@rahulshiraj). https://dev.to/rahulshiraj https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3974321%2Ffe3e391a-bc6a-42b7-b833-aae3b00a1fcc.png https://dev.to/rahulshiraj en rahul kumar Mon, 08 Jun 2026 14:10:17 +0000 https://dev.to/rahulshiraj/how-i-built-an-offline-first-ai-gym-app-in-flutter-with-sqlcipher-gemini-ai-g5h https://dev.to/rahulshiraj/how-i-built-an-offline-first-ai-gym-app-in-flutter-with-sqlcipher-gemini-ai-g5h <blockquote> <h2> The Problem I Was Solving </h2> </blockquote> <p>Most fitness apps sync your health data to the cloud. I wanted an app where everything stays on-device — encrypted and private. No subscriptions, no cloud dependency.</p> <p>So I built <strong>AI Gym Mentor</strong> in Flutter.</p> <h2> The Architecture </h2> <h3> 1. Encrypted Local Database with Drift + SQLCipher </h3> <p>The biggest technical challenge was implementing encrypted local storage. I used <strong>Drift</strong> (formerly Moor) as the ORM and <strong>SQLCipher</strong> for encryption on top of SQLite.</p> <p>This means all workout logs, body measurements, and user data are AES-256 encrypted on-device.</p> <h3> 2. Google Gemini AI Integration </h3> <p>I integrated the <strong>Gemini AI</strong> API to provide real-time coaching. After a workout, the app sends your session data to Gemini and returns personalized feedback, recovery suggestions, and progression advice.</p> <p>The key was keeping the prompt context tight — sending only relevant recent session data rather than the full history.</p> <h3> 3. Riverpod 3.0 State Management </h3> <p>The app uses <strong>Riverpod 3.0</strong> with annotated generators for clean, scalable state management. This was essential given the complexity of syncing Wear OS data, chart state, and AI responses.</p> <h3> 4. Full Tech Stack </h3> <ul> <li>Flutter 3.22+ (iOS, Android, Web)</li> <li>Riverpod 3.0 (Annotated Generators)</li> <li>Drift + SQLCipher (encrypted local DB)</li> <li>Google Gemini AI</li> <li>GoRouter 17.2+</li> <li>Wear OS sync</li> <li>fl_chart for analytics (1RM, volume, body measurements)</li> <li>Lottie + Shimmer animations</li> <li>1,000+ exercise database</li> <li>Multi-language: English, Hindi, Marathi</li> </ul> <h2> Why I Packaged It as a Template </h2> <p>After finishing the app, I realized it had taken months to build all these pieces together correctly. The SQLCipher setup alone took days. The Gemini integration required multiple iterations.</p> <p>So I packaged the complete source code as a developer template — <strong>AI Gym Mentor</strong> — so other developers can build on top of it instead of starting from scratch.</p> <h2> What's Included </h2> <ul> <li>Complete, buildable Flutter source code</li> <li>README with full setup guide</li> <li>Seeded exercise database (exercises.json)</li> <li>.env.example for Gemini API key setup</li> </ul> <p>The template is available here: <a href="https://rahulsaroh.gumroad.com/l/sovgjy" rel="noopener noreferrer">https://rahulsaroh.gumroad.com/l/sovgjy</a></p> <p>Launch price: <strong>$29</strong>. Use code <strong>LAUNCH30</strong> for 30% off (first 10 buyers).</p> <h2> Happy to Discuss </h2> <p>If you have questions about the SQLCipher + Drift setup, the Gemini integration pattern, or the Riverpod architecture — drop a comment below!</p> flutter