At Verizon, I spearheaded a production multi-agent platform that serves 70K+ requests per day with sub-second latency. Its routing layer combines rule-based, semantic, and memory-based strategies.
I also developed Agent Hub, where specialist agents coordinate customer requests, and helped improve first-contact resolution from 62% to 89%. The platform reduced resolution time from 2.5 days to 4 hours.
At Verizon, I automated MLOps incident response with LangGraph agents and improved real-time LLM inference latency by 60%. Before moving into AI engineering, I worked in strategic sourcing at Verizon and Wipro Limited, including contributing to Verizon’s PeopleSoft-to-SAP Ariba migration.
