Skip to main content
Prism

Programming with Confidence in an Uncertain World

Get StartedQuick ReferenceGitHub
example.prism
// AI responses automatically include confidence
analysis = llm("Analyze this code for vulnerabilities")
// Returns: "No critical issues found" ~> 0.85

// Make decisions based on confidence level
uncertain if (analysis) {
  high { deploy_to_production() }    // confidence >= 0.7
  medium { request_human_review() }  // 0.5 <= confidence < 0.7
  low { block_deployment() }         // confidence < 0.5
}

Why Prism?

A programming language designed for the age of AI

First-Class Uncertainty

Attach confidence levels to any value. Uncertainty propagates automatically through calculations.

Built for AI

Native LLM integration with automatic confidence extraction. Handle AI responses with explicit uncertainty tracking.

Confidence Flow Control

Make decisions based on confidence thresholds with intuitive uncertain if statements and branching.

Calibrated Confidence

Extract and calibrate confidence from LLMs using consistency checks, response analysis, and domain-specific patterns.

TypeScript First

Familiar JavaScript-like syntax with full TypeScript support. Modern tooling and IDE integration.

Modular Design

Pick what you need: core language, confidence extraction, LLM providers, and validation tools.

Use Cases

Handle uncertainty in AI responses, sensor data, and complex computations

AI Applications

Handle LLM responses with explicit confidence tracking and make decisions based on reliability thresholds.

IoT & Sensors

Manage noisy sensor data with uncertainty bounds and automatic confidence propagation through pipelines.

Risk Analysis

Make risk-aware decisions with confidence budgets, differential analysis, and threshold gates.

Scientific Computing

Propagate measurement errors and uncertainty through complex calculations automatically.

Get Started in Seconds

npm install @prism-lang/core @prism-lang/llm
Installation Guide →Write Your First Program →