Agentic AI in the Enterprise
How autonomous agents move from demos to governed, production-grade systems.
// engineering · speaking · enterprise AI
Harsh Verma
Principal AI Engineer at Palo Alto Networks · AI speaker on agentic AI, security and trustworthy autonomous systems.
FounderEngineerSpeaker
San Francisco Bay Area
Principal AI Engineer. Enterprise perspective.
Harsh Verma is a Principal Software Engineer in AI at Palo Alto Networks, focused on advancing autonomous agent systems and intelligent AI strategies. A Forbes Technology Council member, speaker, mentor, and advisor, he brings both technical depth and strategic guidance to leaders adopting next-generation AI. He drives responsible, scalable AI adoption — shaping how enterprises innovate, secure, and operationalize intelligent systems.
Harsh’s professional role is Principal Software Engineer in AI at Palo Alto Networks. His engineering work and public talks connect autonomous agents, production machine learning and cybersecurity with the practical requirements of enterprise teams.
Engineering experience
From machine learning to secure agentic AI
Principal Software Engineer in AI | ML · Palo Alto Networks
- Pioneering enterprise AI/ML at the intersection of cybersecurity, big data science and distributed systems.
- Architecting firewall ML models and high-throughput cloud security services.
- Building autonomous security-policy improvement and real-time recommendation engines.
Harsh Verma · AI speaker
Talks for leaders and builders
AI vs AI: The Security Counteroffensive
Defending against adversaries who wield AI — architectures and case studies.
Responsible & Scalable AI Adoption
Strategy for leaders operationalizing AI safely across the organization.
Past talks & recordings
- ACM Sacramento Keynote: When Enterprises Become Multi-Agent Systems
Distinguished Keynote Address titled 'When Enterprises Become Multi-Agent Systems: Secure and Trustworthy Machine Learning & AI for Multi-Domain Applications', analyzing enterprise LLM defense, zero-trust validation, and autonomous agent alignment.
- UC Berkeley SkyDeck Keynote: The Era of Agentic Security
Keynote presentation at UC Berkeley SkyDeck detailing autonomous agent divergence, runtime safety guardrails, prompt provenance, and enterprise cybersecurity architectures.
Enterprise AI, in practice
Explore Harsh’s writing on agentic AI, secure architecture and AI governance, or discuss an enterprise advisory engagement.