diff --git a/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_1.webp b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_1.webp new file mode 100644 index 0000000000..fb8581c6a7 Binary files /dev/null and b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_1.webp differ diff --git a/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_2.webp b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_2.webp new file mode 100644 index 0000000000..e248e7b3e6 Binary files /dev/null and b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_2.webp differ diff --git a/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_3.webp b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_3.webp new file mode 100644 index 0000000000..3a56186f15 Binary files /dev/null and b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_3.webp differ diff --git a/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_4.webp b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_4.webp new file mode 100644 index 0000000000..b81c400bc3 Binary files /dev/null and b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_body_4.webp differ diff --git a/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_logo.webp b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_logo.webp new file mode 100644 index 0000000000..3df01f82f3 Binary files /dev/null and b/sites/www/content/showcase/images/third_party/case_studies/maxint/maxint_logo.webp differ diff --git a/sites/www/content/showcase/maxint.md b/sites/www/content/showcase/maxint.md new file mode 100644 index 0000000000..03abcded3d --- /dev/null +++ b/sites/www/content/showcase/maxint.md @@ -0,0 +1,168 @@ +--- +title: Maxint +description: + Learn how Maxint deployed to six platforms simultaneously with Flutter, saving + 65% in codebase overhead. +headline: Maxint +summary: Building an autonomous financial OS across six platforms with Flutter +appName: Maxint +companyName: Maxint +logo: images/third_party/case_studies/maxint/maxint_logo.webp +card: images/third_party/case_studies/maxint/maxint_logo.webp +locations: + - North America +platforms: + - Mobile + - Desktop + - Web +industries: + - Banking & Finance +successMetrics: + - perc: 65 + desc: code avoided vs native multi-platform + - text: "3-4x" + desc: fewer front-end engineers needed + - text: "6" + desc: platforms launched from day one +tags: + - mobile + - desktop + - web + - ios + - android + - macos + - windows + - linux + - ai +publishDate: 2026-08-21 +--- + +[Maxint](https://maxint.com) is an autonomous financial operating system. Built +with Flutter, it unifies enterprise accounting, multi-platform monetization, and +AI-powered robo-advising into a single, precision-engineered workspace designed +to provide the signal minus the noise. + +As a lean startup, Maxint needed to rapidly deploy an enterprise-grade financial +operating system across iOS, Android, macOS, Windows, Linux, and Web +simultaneously. By launching on every major app store from day one, the team +leveraged organic search discovery across multiple platform ecosystems to drive +consistent user growth without spending a dollar on paid marketing. + +Maxint running across desktop, tablet, and mobile platforms + +**Trial by fire: From hackathon to production in three months** + +Rather than spending weeks evaluating frameworks in committee, the team tested +Flutter during the TechCrunch Disrupt hackathon. Faced with tight time +constraints after evaluating options like React Native, Ionic, Xamarin, and +native platforms, they chose Flutter to build a fully functional proof of +concept for Maxint in less than 24 hours. + +The success of this hackathon validated Flutter's developer velocity, responsive +charting, and local state management across devices. Following the hackathon +success, a lean team spent three months architecting the production-ready +financial OS, deploying native applications across all six target platforms +simultaneously. This was a timeline that likely would have required over a year +using siloed native teams. + +**Why Flutter? Single codebase portability and rendering performance** + +Portability across desktop and mobile served as Maxint's foundational growth +engine. By maintaining a single Dart codebase, complex mission-critical +logic—such as their double-entry ledger, local encryption, and financial +reporting engine—only had to be written and audited once, guaranteeing complete +feature parity and data accuracy across all user devices. + +To meet the high UI performance standards of financial software, Maxint relies +on Flutter's Impeller rendering engine on iOS and Android. By pre-compiling +shaders ahead-of-time (AOT) and using modern native graphics APIs like Metal and +Vulkan, Impeller renders complex, data-dense financial charts at 120 frames per +second without blocking the UI thread. Additionally, Flutter's hot reload +feature accelerated iteration, allowing the team to refine dense ledger layouts +and state logic without restarting the app. + +Maxint enterprise accounting interface rendering at 120fps + +**Building Finsight: Bidirectional conversational AI with Flutter** + +Maxint built Finsight, a fluid voice-driven financial assistant, directly within +the Flutter ecosystem by orchestrating +[`speech_to_text`](https://pub.dev/packages/speech_to_text) and +[`flutter_tts`](https://pub.dev/packages/flutter_tts). The application uses +continuous listening and partial transcription to stream user queries to +language models while the user speaks, minimizing perceived latency. + +When generating spoken feedback, Finsight attaches context tags (such as urgent +alerts or routine portfolio summaries) to dynamically adjust parameters like +`setSpeechRate` and `setPitch` in `flutter_tts`. Using +[Riverpod](https://riverpod.dev/) for state management, the on-screen visualizer +pulses in synchronization with audio callbacks, maintaining a unified +interactive state across mobile, desktop, and web platforms. + +Maxint complex market analysis powered by local LLMs + +**Overcoming multi-platform friction: Building Orca** + +Synchronizing user entitlements across the Apple App Store, Google Play, +Microsoft Store, and Stripe presented a major technical hurdle due to isolated +store silos. To solve multi-platform subscription management, Maxint developed +an internal unified billing infrastructure and released it to the developer +community as [Orca](https://orca.maxint.com/). + +Building for desktop also required investing in desktop-native UX paradigms, +including information-dense layouts, robust keyboard shortcuts, right-click +context menus, and window state management, while keeping double-entry ledger +math completely isolated from UI rendering. + +**Key results and business impact** + +Building with Flutter transformed Maxint's resource efficiency and release +cadence: + +- **65% less code:** Avoided writing and maintaining at least 65% more code + compared to managing separate native codebases (Swift, Kotlin, C\#, C++, and + TypeScript). +- **3x to 4x team efficiency:** Operates with 3x to 4x fewer front-end engineers + than needed for native multi-platform parity. +- **Day-one six-platform cadence:** Pushes simultaneous, feature-complete + updates to iOS, Android, Web, macOS, Windows, and Linux on a weekly basis. +- **Ecosystem contribution:** Created and launched Orca to streamline + cross-platform billing for Flutter developers. + +
+
+ +**What's next for Maxint and Orca** + +Maxint's roadmap includes expanding Orca's SDK capabilities across web and +desktop platforms for unified paywalls and revenue analytics. For the core +platform, Maxint is building upon its current on-device receipt parsing, powered +by Google’s ML Kit, by doubling down on privacy-first AI intelligence. + +Unified cross-platform monetization architecture powered by Orca + +The team is integrating small-footprint local LLMs (including custom +135M-parameter transformer models) to execute highly advanced document parsing, +cashflow forecasting, and transaction categorization entirely on the user's +hardware. The team is also expanding their tax engine to support +multi-jurisdiction calculations, local currency reconciliation, and native +electronic tax filing with the IRS. + +
+