Introduction

Bridgic is an innovative programming framework designed to create agentic systems, ranging from deterministic workflows to autonomous agents. It introduces a new paradigm that simplifies the development of agentic systems.

Core Features¶

• Orchestration: Bridgic helps manage the execution flow of your AI applications by leveraging both predefined dependencies and dynamic routing.
• Parameter Binding: There are three ways to pass data among workers—Arguments Mapping, Arguments Injection, and Inputs Propagation—thereby eliminating the complexity of global state management.
• Dynamic Routing: Bridgic enables conditional branching and intelligent decision-making through an easy-to-use ferry_to() API that adapts to runtime dynamics.
• Dynamic Topology: The topology can be changed at runtime in Bridgic to support highly autonomous AI applications.
• Modularity: In Bridgic, a complex agentic system can be composed by reusing components through hierarchical nesting.
• Human-in-the-Loop: A Bridgic-style agentic system can request feedback from human whenever needed to dynamically adjust its execution logic.
• Serialization: Bridgic employs a scalable serialization and deserialization mechanism to achieve state persistence and recovery, enabling human-in-the-loop in long-running AI systems.
• Systematic Integration: A wide range of tools and LLMs can be seamlessly integrated into the Bridgic world, in a systematic way.
• Customization: What Bridgic provides is not a "black box" approach. You have full control over every aspect of your AI applications, such as prompts, context windows, the control flow, and more.