Skip to content
Dana Logo
Agent-native programming language and runtime
powered by OpenDXA

Dana — The Agent-Native Evolution of AI Development

Beyond AI coding assistants: Write agents that learn, adapt, and improve themselves in production

Brought to you by Aitomatic and other AI Alliance members.

What if your code could learn, adapt, and improve itself in production—without you?

Dana bridges the gap between AI coding assistance and autonomous agents through agent-native programming: native agent primitives, context-aware reason() calls that adapt output types automatically, self-improving pipelines with compositional | operators, and functions that evolve through POET feedback loops.

Documentation: Choose Your Path

I want to build with Dana

→ For Engineers - Practical guides, recipes, and references Perfect for developers who want to get working quickly

What you'll find: - 5-minute setup and first agent tutorial - Complete Dana language reference and REPL guide - Real-world recipes for chatbots, document processing, and workflows - Troubleshooting guides and error references

Start here: Getting Started

I'm evaluating Dana for my team

→ For Evaluators - Comparisons, ROI analysis, and proof of concepts Perfect for technical leads and decision makers

What you'll find: - ROI calculator and competitive analysis - Risk assessment and technical evaluation frameworks - Proof of concept guides and adoption strategies - Decision frameworks and implementation roadmaps

Start here: Evaluation Guide

I want to contribute or extend Dana

→ For Contributors - Architecture, codebase, and development guides Perfect for developers who want to modify or extend the system

What you'll find: - Complete architecture deep dive and codebase navigation - Development environment setup and contribution guidelines - Extension development for capabilities and resources - Testing frameworks and documentation standards

Start here: Development Setup

I want to understand the philosophy and theory

→ For Researchers - Manifesto, theory, and academic context Perfect for researchers and those interested in the theoretical foundations

What you'll find: - Dana manifesto and neurosymbolic computing foundations - Research opportunities and collaboration frameworks - Theoretical analysis and future research directions - Academic partnerships and publication opportunities

Start here: Research Overview

I'm interested in investment opportunities

→ For Investors - The agent-native evolution of AI development For accredited investors and principals only (no agents or intermediaries)

What you'll find: - How Dana captures the convergence of AI coding assistants and autonomous agents - Market opportunity at the intersection of two validated $B+ markets - Agent-native programming advantages over retrofitted frameworks - Production validation across enterprise deployments

Contact: investors@aitomatic.com

Why Dana?

Dana transforms AI development from brittle, unpredictable systems to reliable, auditable automations through agent-native architecture:

  • 🔍 Transparent: Every step is visible and debuggable through imperative programming
  • 🛡️ Reliable: Built-in verification and error correction with structured state management
  • ⚡ Fast: 10x faster development cycles with clear control flow
  • 🤖 Agent-Native: Purpose-built for multi-agent systems with first-class agent primitives
  • 🧠 Context-Aware: reason() calls that adapt output types automatically based on usage
  • 🔄 Self-Improving: Functions that learn and optimize through POET in production
  • 🤝 Collaborative: Share and reuse working solutions across domains
  • 🌐 Domain-Expert: Seamless integration of specialized knowledge and expertise

Core Innovation: Agent-Native Programming

Dana provides an agent-native imperative programming model that bridges development assistance with autonomous execution:

# Traditional AI: Opaque, brittle
result = llm_call("analyze data", context=data)

# Dana: Transparent, self-correcting with explicit state management
analysis = reason("analyze data", context=data) # Auto-scoped to local (preferred)
while confidence(analysis) < high_confidence:
 analysis = reason("refine analysis", context=[data, analysis])

# Clear state transitions and auditable reasoning
public:result = analysis
use("tools.report.generate", input=public:result)

Context-Aware Intelligence: Same reasoning, different output types based on usage: 

risk_score: float = reason("assess portfolio risk", context=portfolio)
risk_details: dict = reason("assess portfolio risk", context=portfolio) 
risk_report: str = reason("assess portfolio risk", context=portfolio)

Self-Improving Pipelines: Compositional operations that optimize themselves: 

portfolio | risk_assessment | recommendation_engine | reporting  # Gets smarter via POET

Agent-Native Programming: Write agents as first-class primitives: 

agent FinancialAnalyst:
    def assess_portfolio(self, data):
        return reason("analyze risk factors", context=data)  # Function learns over time

Quick Navigation by Use Case

Building AI Agents

📊 Business Evaluation

🔬 Research & Development

Platform Extension

🌟 Success Stories

"I used to spend hours debugging prompt chains and patching brittle scripts. Every new document or edge case meant another late night. With Dana, I finally feel in control. My automations are clear, reliable, and easy to improve. What used to take our team weeks now takes days or even hours."

— Sarah K., Lead AI Engineer at FinTech Solutions

"Dana's transparency was a game-changer for our compliance requirements. We can audit every decision, understand every step, and trust our AI systems in production. The ROI was evident within the first month."

— Michael R., CTO at Healthcare Analytics

🚦 Getting Started Paths

⚡ 5-Minute Demo

pip install opendxa
opendxa demo chatbot
Try the demo →

Learn the Concepts

  1. What makes Dana different
  2. Dana language basics
  3. Agent architecture overview
  4. DANKE knowledge engine

Solve Your Use Case

📞 Community & Support

💬 Get Help

🤝 Get Involved

🏢 Enterprise Support

📖 Documentation Structure

This documentation is organized by audience with cross-references and maintained through structured AI-assisted processes:

docs/
├── for-engineers/ # Practical development guides
│ ├── setup/ # Installation and configuration
│ ├── recipes/ # Real-world examples and patterns
│ ├── reference/ # Language and API documentation
│ └── troubleshooting/ # Common issues and solutions
├── for-evaluators/ # Business and technical evaluation
│ ├── comparison/ # Competitive analysis and positioning
│ ├── roi-analysis/ # Cost-benefit and ROI calculations
│ ├── proof-of-concept/ # Evaluation and testing guides
│ └── adoption-guide/ # Implementation and change management
├── for-contributors/ # Development and extension guides
│ ├── architecture/ # System design and implementation
│ ├── codebase/ # Code navigation and understanding
│ ├── extending/ # Building capabilities and resources
│ └── development/ # Contribution and testing guidelines
├── for-researchers/ # Theoretical and academic content
│ ├── manifesto/ # Vision and philosophical foundations
│ ├── neurosymbolic/ # Technical and theoretical analysis
│ ├── research/ # Research opportunities and collaboration
│ └── future-work/ # Roadmap and future directions
├── archive/ # Preserved original documentation
│ ├── original-dana/ # Authoritative Dana language specification
│ ├── original-core-concepts/ # Original architectural concepts
│ └── original-architecture/ # Historical system design
├── internal/ # Internal planning and requirements
└── .ai-only/ # AI assistant structured references
 ├── documentation.md # Documentation maintenance prompts
 ├── documentation-maintenance.md # Structured update procedures
 ├── project.md # Project structure guide
 ├── opendxa.md # System overview and components
 ├── dana.md # Dana language technical reference
 └── functions.md # Function catalog and registry

Documentation Maintenance

This documentation is maintained through structured processes that ensure: - Function Registry: Automated tracking of new Dana functions and capabilities - Example Validation: Regular testing of all code examples with current syntax - Content Gap Analysis: Weekly assessment of documentation coverage - Cross-Audience Updates: Synchronized updates across all audience trees - AI-Assisted Quality: Structured prompts for consistent maintenance

The .ai-only/ directory contains reference materials and maintenance procedures that keep this documentation current and comprehensive.

Ready to transform your AI development? Choose your path above and start building transparent, reliable AI automations with Dana.

Copyright © 2025 Aitomatic, Inc. Licensed under the MIT License.
https://aitomatic.com