The General-Purpose AI Agent is a smart AI routing system that optimizes cost, speed, and accuracy by dynamically selecting the best model for each task. It supports eight capabilities, including factual question answering, mathematical reasoning, sentiment analysis, text summarization, named entity recognition, code debugging, logical reasoning, and code generation. A lightweight rule-based classifier first identifies the task category and routes simple NLP tasks to a local Ollama model for zero API cost, while medium and complex tasks are handled by Fireworks AI models with automatic fallback when needed. The architecture minimizes token usage through intelligent routing, optimized prompts, and capped output lengths. Fully containerized with Docker and configurable via environment variables, the agent is portable, scalable, and designed for efficient, reliable execution in the AMD Hackathon environment.
Category tags: