At Green Circle for Cybersecurity, I mapped business workflows and designed and tested LLM, RAG, and AI-agent workflows with validation, escalation, and human review. I also evaluated AWS Bedrock RAG experiments using S3-backed knowledge sources and citation-grounded test cases.
I co-developed GreenLead for sales intelligence and AI automation, and built an Arabic/English RAG benchmark to evaluate retrieval grounding, citations, and answer quality. My projects also include an academic-advising chatbot and a typed LangGraph agent router.

