Experience

• Co-designed and built Syntropix DB, an AI-native ERP platform from the ground up, enabling first-class collaboration with LLM agents instead of retrofitting legacy ERP architectures.
• Architected a modular, plugin-based system to support rapid customization for small and mid-sized businesses with non-standard workflows.
• Worked directly with customers alongside the CEO, acting as a technical bridge to translate business workflows and edge cases into concrete system and architectural requirements.
• Collaborated in a lean two-engineer team, emphasizing UX-driven simplification and long-term code maintainability through refactoring, critical reviews, and consistent design patterns.

• Core developer of the open-source LLM multi-agent framework CAMEL (15.1K stars).
• Led the design and implementation of Workforce, an agent orchestration system that has become one of the core architectures within CAMEL for coordinating multi-agent task execution.
• Refactored the core agent execution pipeline by replacing patch-based conditional logic with a polymorphic design, encapsulating agent-specific behaviors within agents and simplifying the shared runtime.
• Reduced core runtime code by over one-third (from ~1500 LOC to under 1000 LOC), significantly improving readability, maintainability, and the ease of extending new agent types without modifying shared execution logic.

M.S. in Computer Science

B.E. in Computer Science and Technology