ESAA: Event Sourcing for Autonomous Agents in LLM-Based Software Engineering

Researchers have introduced ESAA (Event Sourcing for Autonomous Agents), a new architecture that improves LLM-based autonomous agents by separating cognitive intention from state mutation using structured JSON events and deterministic orchestration. The system addresses key limitations like context degradation and execution reliability, with successful validation through multi-agent case studies using various LLMs including Claude Sonnet and GPT-5.

• →ESAA architecture separates agent cognitive processes from project state changes to improve reliability and traceability.
• →The system uses structured JSON events stored in append-only logs with deterministic orchestration and verification mechanisms.
• →Multi-agent case studies demonstrated successful concurrent orchestration across heterogeneous LLMs including Claude Sonnet 4.6 and Codex GPT-5.
• →The architecture addresses critical LLM agent limitations like context degradation and the gap between probabilistic generation and deterministic execution.
• →Both single-agent and multi-agent implementations achieved successful completion with proper verification status.