Quadesto is an AI-powered financial data visualization platform built for finance professionals who need correct methodology in their charts. We exist because financial visualizations are not generic charts with financial data overlaid — they require domain-specific computation, calibration, and interpolation that general-purpose tools do not implement.
Quadesto connects to your data sources — APIs, databases, CSV files, or direct paste — and uses AI to analyze the structure and type of your data. From there, the platform computes derived metrics (spreads, moving averages, Greeks, forward rates), selects the correct chart type and methodology, and renders publication-ready visualizations. Every chart can be refined through natural language, exported as PNG or PDF, or embedded on any website with a single line of code.
Financial charts require domain-specific methodology that general-purpose charting tools do not provide. A yield curve is not a line chart — it requires monotone convex interpolation to avoid arbitrage-violating oscillations. A volatility surface requires SVI calibration to produce a no-arbitrage parameterization across strikes and tenors. Options analytics require Black-Scholes pricing, Greeks computation, and implied volatility extraction.
Before Quadesto, finance professionals had two options: use a general-purpose tool that gets the methodology wrong, or write custom code — which costs weeks of development per chart type and produces output that is difficult to maintain, share, or embed.
Quadesto gives finance professionals publication-ready charts with correct methodology in minutes, not weeks. The platform handles the computation, calibration, and rendering so the user can focus on analysis and communication.
Quadesto stores user credentials (API keys, database connection strings) in an AES-256-GCM encrypted vault with per-credential random initialization vectors. All database tables enforce Postgres row-level security policies to prevent cross-tenant data access. Server-side URL fetching validates against private IP ranges and DNS rebinding to prevent SSRF attacks. The platform's security architecture is aligned with SOC2 Type II trust service criteria.
For a detailed breakdown of our security controls, see the security page.
Quadesto is built on Next.js and TypeScript with ECharts for rendering, Supabase for authentication and storage, and Anthropic Claude for AI-powered data analysis and chart generation. The platform uses pgvector to store and retrieve financial methodology documentation, ensuring that AI recommendations are grounded in domain-specific knowledge rather than generic pattern matching.
Quadesto is not a black box. The computation engine exposes 90+ registered functions — from moving averages and Bollinger Bands to Black-Scholes pricing and yield curve bootstrapping — each with documented inputs, outputs, and methodology. Users can inspect every derived value and understand exactly how it was computed.