ORBIT

Organizational Relationship & Board Interconnection Tracker

Comment 1

Inspiration

In today's interconnected global economy, understanding the complex web of corporate relationships is crucial for investors, analysts, and researchers. Traditional financial tools show you data in isolation, stock prices, earnings reports, and news, but fail to reveal the hidden connections that truly drive market dynamics.

We were inspired by the 2008 financial crisis, where hidden interconnections between institutions amplified risk across the entire system. When Lehman Brothers collapsed, the ripple effects weren't obvious until it was too late. What if we could visualize these connections before crisis hits?

ORBIT was born from a simple question: What would it look like if you could see the entire corporate universe, every ownership stake, every supply chain dependency, every board interlock, as an interactive 3D constellation?

What it does

ORBIT (Organizational Relationships & Business Intelligence Tool) is an immersive 3D visualization platform that maps the complex network of relationships between S&P 500 companies.

3D Knowledge Graph - Navigate through 500+ companies rendered as stars in a cosmic visualization, with 10 distinct relationship types shown as color-coded edges: Ownership, Partnership, Supplier, Client, Creditor, Debtor, Joint Ventures, Licensing, Swaps, and Board Interlock.

Path Finder - Discover hidden connections between any two companies with configurable depth (2-7 hops). See how Apple connects to Boeing through unexpected intermediaries.

AI-Powered News Analysis - Paste a news article and watch as Gemini AI identifies affected companies and highlights them on the graph with bullish/bearish indicators.

Relationship Deep Dives - Click any edge to see detailed SEC filings, financial data, supply chain tiers, and related news articles.

Watchlist - Track your portfolio companies and filter the graph to show only your investments and their connections.

How we built it

Frontend: Next.js 14 with App Router and TypeScript for type-safe development

3D Rendering: Three.js via react-force-graph-3d for immersive graph visualization

Animation: Framer Motion for fluid UI transitions and micro-interactions

AI Integration: Google Gemini 2.0 Flash for NLP, entity extraction, and sentiment analysis

Styling: Tailwind CSS combined with custom CSS-in-JS for the cosmic aesthetic

Data Layer: Custom S&P 500 dataset with 500+ nodes and 1000+ relationship edges across 10 categories

We used a force-directed physics simulation so related companies naturally cluster together, making patterns visible at a glance. Edge curvature algorithms ensure multiple relationships between the same companies remain distinguishable.

Challenges we ran into

WebGL Performance - Rendering 500+ 3D nodes with glow effects and 1000+ edges pushed browser limits. We encountered WebGL context loss crashes and had to optimize particle counts, implement graceful fallbacks, and use fixed-size rendering techniques.

Multi-Edge Visualization - When two companies have multiple relationships (e.g., Microsoft is both client AND partner of OpenAI), edges would overlap. We solved this with dynamic curvature calculation that spreads edges symmetrically.

AI Entity Extraction - Gemini sometimes returned malformed JSON or company names not in our database. We built multi-layer validation with aggressive JSON cleanup and fuzzy matching against our S&P 500 dataset.

Real-time Filtering - Toggling relationship filters on 1000+ edges caused lag. We optimized with careful useMemo dependencies and avoided recreating large data structures on each render.

Accomplishments that we're proud of

We created a stunning cosmic visual design that makes financial data feel approachable and engaging, companies literally glow when you interact with them.

Our path-finding algorithm discovers all connections between any two companies in under 100ms, even at 7-hop depth.

The AI news analysis correctly identifies 85%+ of company mentions and appropriately classifies bullish/bearish sentiment.

We built an elegant curved-edge solution that clearly shows when companies have multiple relationship types without visual clutter.

The entire UI uses a mobile-inspired bottom sheet pattern with smooth collapse/expand animations.

What we learned

Three.js at Scale - Managing GPU memory in WebGL applications requires careful geometry disposal and texture management to prevent crashes.

Graph Theory in Practice - Implementing BFS/DFS for path finding and understanding why force-directed layouts naturally reveal community structure.

LLM Prompt Engineering - Crafting prompts that reliably return structured JSON, including fallback strategies when AI responses are malformed.

TypeScript for Complex State - Managing interconnected state (selected nodes, highlighted paths, enabled filters, watchlist) benefits enormously from strict typing.

UX for Dense Data - Showing 1000+ relationships without overwhelming users requires progressive disclosure and smart defaults.

What's next for ORBIT

Real-time Market Data - Integrate live stock prices and show node size/color based on market cap or daily performance.

Live News Ingestion - Continuous news feed with automatic graph updates as new relationships are discovered.

Alert System - Notifications when watchlist companies have relationship changes or are mentioned in significant news.

Global Expansion - Extend beyond S&P 500 to international markets including FTSE, Nikkei, and DAX.

Quantitative Analysis - Calculate network centrality scores to identify systemically important companies before crises emerge.

ORBIT transforms abstract financial relationships into an explorable universe. Because in today's market, it's not just what you know, it's how everything connects.

Built With

+ 19 more
Share this project:

Updates