Multi Agent Claude Debugging and Optimization for SaaS Application

We are looking for an experienced developer to debug and fix an existing multi agent AI system built with Claude for our SaaS application. The main challenges involve Python code errors, incorrect API requests, agent execution issues, and problems parsing AI generated responses. Agents are not consistently processing requests or returning the expected output, affecting the application's functionality. The objective is to identify the root causes, fix the existing implementation, ensure requests are correctly constructed and processed, resolve response parsing issues, and improve the reliability of the overall agent workflow. Required Skills: Python, SQL, Claude API, multi agent AI workflows, JSON parsing, Microsoft Azure, and Databricks. Experience debugging existing AI applications and troubleshooting LLM API integrations is highly preferred.