Supernova

Product Introduction

1. Definition: Supernova is a unified data connectivity and AI analytics platform. Technically, it is an encrypted data lakehouse built on Apache Iceberg, integrated with a real-time sync engine and a type-safe SQL query layer, designed to connect directly to frontier AI models like Anthropic's Claude and OpenAI's Codex via the Model Context Protocol (MCP).
2. Core Value Proposition: It exists to eliminate the data access bottleneck for startups and technical teams. Supernova syncs live data from business applications (like Stripe, HubSpot, PostgreSQL) into a single, queryable data lake, enabling anyone to perform complex data analysis, investigate performance, and ask business questions directly within AI tools without requiring engineering resources or a traditional Business Intelligence (BI) stack.

Main Features

1. Unified Data Lakehouse with Real-Time Sync: Supernova acts as a managed data pipeline and storage layer. It connects to over 30+ sources (e.g., Postgres, MongoDB, Stripe, HubSpot) via FTL connectors, syncing data into an encrypted Apache Iceberg data lake in object storage. How it works: Connectors perform initial historical syncs and then maintain real-time or scheduled incremental updates, tracking full schema history. The Iceberg format ensures open, vendor-neutral storage that supports "time travel" for data versioning.
2. Native AI Integration via MCP (Model Context Protocol): This is the core interface for analysis. Users connect the endpoint mcp.supernova.ai/mcp to Claude Desktop or Codex. Once connected, the AI model has direct, schema-aware access to the entire synchronized data lake. Users can ask natural language questions (e.g., "Which customers churned last quarter, and why?") and the AI generates and executes the correct TypeSQL queries, returning contextual, analyzed answers.
3. TypeSQL: A Type-Safe SQL Dialect: Supernova's query engine uses TypeSQL, a schema-aware SQL language that provides autocomplete and type-checking. It infers column types (e.g., currencyAmount, email, geoPoint) and validates joins and syntax before execution. This prevents runtime errors (e.g., catching a typo like stauts instead of status) and enhances developer productivity, functioning like a TypeScript layer for SQL.
4. Particle: Embedded Frontier AI Models: Beyond MCP connectivity, Supernova includes its own AI chat interface called Particle. Users can describe dashboards or ask questions in plain English, and Particle's integrated AI models will generate the necessary data models, SQL queries, and visualizations, making self-service analytics accessible without SQL knowledge.
5. Comprehensive CLI & GitOps Workflow: The Supernova Command Line Interface (CLI) is a single binary tool for full platform management. Users can connect data sources (supernova connect stripe), run queries, tail real-time database changes, and configure MCP. Furthermore, all data models and dashboards are managed via Git, enabling version control, code review, and CI/CD integration for analytics workflows.

Problems Solved

1. Pain Point: The high latency and dependency on data engineers or analysts for business insights. Traditional workflows require moving data to a warehouse, modeling it, and building dashboards, which is slow and creates bottlenecks.
2. Target Audience: Startup founders, product managers, operations leads, and go-to-market teams in tech companies who need immediate data access but lack a dedicated data team. It also serves full-stack and data engineers who want to streamline the data pipeline and provide instant analytics capabilities.
3. Use Cases:
   • CEO/Founder Performance Investigation: Instantly querying combined Stripe and HubSpot data to understand revenue drivers and customer churn reasons without a weekly report.
   • Product Manager Usage Analysis: Analyzing user behavior data from PostgreSQL and Mixpanel to identify feature adoption trends and drop-off points.
   • Marketing Operations Dashboard Creation: Using Particle AI to build a live dashboard of signups by week and campaign ROI by connecting Google Ads, Salesforce, and Shopify.
   • Engineer-Free Business Reporting: Enabling the finance team to run complex cohort retention analysis directly by asking Claude a question, bypassing SQL ticket queues.

Unique Advantages

1. Differentiation: Unlike traditional BI tools (Tableau, Looker) that require a pre-built warehouse and semantic layer, Supernova provides the pipeline, lake, and AI-native query engine in one product. Unlike simple dashboard tools, it offers deep, ad-hoc investigative power. Unlike other MCP servers, it is a full-stack data platform, not just a connector.
2. Key Innovation: The deep integration of a type-safe SQL engine (TypeSQL) with a unified data lakehouse, exposed directly to frontier AI models via MCP. This creates a closed loop where natural language is translated into validated, executable SQL against live, typed data, with results formatted contextually (e.g., currency as $48,120.00, geo-points as interactive maps). The "zero-copy" architecture using open Iceberg tables also prevents vendor lock-in.

Frequently Asked Questions (FAQ)

1. How does Supernova's pricing compare to a traditional data stack? Supernova uses a transparent consumption-based model (e.g., $0.15/GB-hour for compute, $0.05/GB-month for storage). This often results in significant cost savings compared to the combined fees of separate ETL tools, cloud data warehouses (like Snowflake), BI platform licenses, and per-seat costs, especially for startups with variable usage.
2. Is my data locked into Supernova? No. Your data resides in your own encrypted Apache Iceberg tables in object storage (e.g., AWS S3). The Iceberg format is an open standard. You can point other compatible query engines (like Spark, Trino, or your existing warehouse) directly at these tables for "zero-copy" access, ensuring full portability and no lock-in.
3. Do I need a data warehouse to use Supernova? No. Supernova is an all-in-one platform that replaces the need for a separate ETL pipeline and a cloud data warehouse for many use cases. It includes the sync engine (pipeline), the managed Iceberg lake (storage), and the TypeSQL query engine. If you have an existing warehouse, it can read from the Supernova lake directly.
4. How secure is the connection between Claude and my company data? Data flows through a secure MCP connection to mcp.supernova.ai. Your source data credentials are never exposed to the AI model. The AI sends query requests to Supernova, which executes them against the data lake and returns only the result set, maintaining a separation between the AI's reasoning and your raw data storage.
5. What if Supernova doesn't have a connector for my data source? The platform offers connectors for 30+ common SaaS apps and databases. For custom sources, Supernova offers to build a dedicated FTL connector within 24 hours, demonstrating a commitment to comprehensive data connectivity.