FundPilot

Smart Tracking for Smarter Investing.

Comment

Login Details to access the app

email: [email protected] password: Password123@ You can also create an account

Inspiration

Wealth is no longer stored in a single vault. Today, an average investor’s financial life is scattered: equities on a primary broker, crypto in a hardware wallet, commodities tracked elsewhere, and perhaps a real estate investment or private bond documented only in a dusty spreadsheet. This creates The Fragmentation Gap. Information is siloed across dashboards and statements that were never built to communicate with each other. Simple, critical questions like “What is my actual total exposure?” or “Is my portfolio truly diversified?” become nearly impossible to answer without hours of manual data entry. FundPilot was born from a simple vision: to build a single, intelligent control centre. We wanted to give investors a clear lens into their entire financial world, harmonising publicly traded securities with privately held instruments into one real-time source of truth.

What it does

FundPilot acts as the centralised intelligence layer for modern portfolios. Unified Portfolio View: Seamlessly aggregate stocks, crypto, commodities, and fixed-income instruments into a single, high-fidelity dashboard. Real-Time Valuation: For listed assets, the system continuously syncs with global markets to reflect current values and performance deltas. Support for the Non-Traditional: Manual tracking for bonds, private structured products, or real estate ensures that an investor’s "net worth" is actually accurate, not just a partial snapshot. AI-Driven Risk Insights: Using Google Gemini, the platform moves beyond raw numbers. It analyses portfolio composition to detect sector imbalances or geographic overexposure, offering actionable narrative intelligence. Proactive Guardrails: Set loss thresholds and receive daily AI-generated performance summaries, allowing investors to stay informed without being glued to the screen.

How we built it

We engineered FundPilot with a modular, performance-first architecture to ensure financial data remains reliable and responsive. Frontend: Built with Flutter for a premium, cross-platform mobile experience, using Riverpod for robust, reactive state management. Backend: A high-concurrency Go microservice (using the Fiber framework) handles complex data ingestion, price synchronisation, and background workers. Database: PostgreSQL with GORM manages structured persistence, maintaining historical snapshots for performance tracking. AI Intelligence: Google Gemini Pro powers the core analysis engine, translating rows of data into human-readable advice. Real-Time Alerts: Firebase Cloud Messaging (FCM) delivers instant alerts based on server-side triggers. The Mathematics of Performance FundPilot continuously evaluates portfolios using standardized financial models:

Total Portfolio Valuation (V_p): $$ V_p = \sum_{i=1}^{n} (Q_i \cdot P_i) $$

Dynamic Asset Weighting (w_i): $$ w_i = \frac{V_i}{V_p} $$

Performance Delta (\Delta %): $$ \Delta% = \left( \frac{V_{now} - V_{initial}}{V_{initial}} \right) \cdot 100 $$

Challenges we ran into

The Market Sync Paradox: Syncing pricing data for dozens of asset types without triggering API rate limits required building a sophisticated scheduler that intelligently batches requests based on market volatility. Race Conditions at Scale: Early builds faced initialisation conflicts with mobile permissions and notification triggers. I solved this by implementing a strictly sequenced startup registry to serialise tasks safely. AI Structure Enforcement: Getting an LLM to consistently output structured financial insights instead of generic advice required rigorous prompt engineering and JSON-schema enforcement at the API layer.

Accomplishments that we're proud of

Architectural Harmony: How to build a system where a Dart frontend and a Go backend work in perfect sync using REST and Cloud Messaging. Reactive UX: The importance of optimistic UI updates and reactive state management in keeping financial dashboards feeling "alive." Production-Ready Security: Best practices for managing credentials and sensitive data in a modern cloud-native deployment. Reactive Precision: We successfully built an interface that reflects valuation changes across the entire stack in near real-time, delivering a "live" feel to a heavy financial tool. Unified Entitlement Security: Integrating a custom Go backend with RevenueCat to enforce live, fail-closed subscription security across mobile and web layers.

What we learned

Architectural Harmony: How to build a system where a Dart frontend and a Go backend work in perfect sync using REST and Cloud Messaging. Reactive UX: The importance of optimistic UI updates and reactive state management in keeping financial dashboards feeling "alive." Production-Ready Security: Best practices for managing credentials and sensitive data in a modern cloud-native deployment. Building for finance is a lesson in trust and clarity. I learned that infrastructure decisions, like choosing Go for the backend scheduler or Riverpod for the frontend, directly impact a user’s confidence in the data. Most importantly, we saw how a thoughtful system can take "the noise" of fragmented data and turn it into the "signal" of actionable insight.

What's next for FundPilot

The journey is just beginning. Our roadmap includes: Automated Bank Syncing: Integrating with Open Banking APIs (like Plaid) to automate the entry of cash and savings accounts. Predictive Analytics: Using AI not just to analyse the present, but to simulate future market scenarios and their impact on the portfolio. Community Insights: Anonymous, aggregated sentiment analysis to show users how other "FundPilots" are diversifying their portfolios.

Built With

Share this project:

Updates

Submission history