Mr. Pradeep Kiran Veeravalli

Agentic AI in Action: Transforming Enterprise Customer Support at Scale

Abstract:

Agentic Artificial Intelligence is reshaping enterprise customer support by combining autonomous decision-making, contextual reasoning, and goal-directed behavior to deliver measurable gains in speed, cost, and satisfaction. This talk presents a real-world deployment where an agentic AI system handled high inquiry volumes across diverse issue categories and multiple languages while maintaining near-continuous availability. The system achieved high independent resolution rates for tier‑1 inquiries, reduced average time to first response from hours to minutes, and delivered substantial operational cost savings—all while improving CSAT and time-to-resolution. We unpack the architectural pillars that enabled these outcomes: hybrid decision frameworks blending deterministic safeguards with adaptive learning; deep integration across CRM, knowledge, and operational systems for 360° context; and scalable, cloud-native services that maintained low-latency performance under concurrent load. We also address critical success factors often underestimated in practice, including data preparation at scale, legacy integration complexity, and change management for human‑AI collaboration. Finally, we detail privacy-by-design controls—sensitivity-aware routing, graduated access, and options for private inference for protected data—that ensured compliance without sacrificing performance. Attendees will leave with a pragmatic blueprint to implement agentic AI in customer support, from pilot to production, with governance that scales across regulated environments.

Profile:

Pradeep Kiran Veeravalli is a Lead Member of Technical Staff at Salesforce and a researcher-practitioner specializing in applied AI and scalable cloud solutions for service and customer operations. He designs agentic AI systems that combine deterministic safeguards with adaptive learning to deliver reliable automation, measurable ROI, and privacy-by-design. At Salesforce, Pradeep builds production-grade architectures that integrate deeply with CRM, knowledge, and operational platforms—enabling low-latency, multilingual support at scale with strong observability and governance. He advises teams on integration strategy, data quality, and progressive autonomy rollouts that balance speed with safety and compliance. His interests include bounded autonomy frameworks, sensitivity-aware routing for regulated data, and performance engineering for large-scale NLP. Bridging industry and research, Pradeep speaks and writes about moving from proofs-of-concept to resilient, governed AI systems that improve time-to-resolution, elevate customer satisfaction, and reduce cost-to-serve while preserving brand voice and regulatory requirements.