Unlock how Salesforce architects can design secure, trusted, next-gen agentic AI and Agentforce solutions. You will walk away with five battle-tested design pillars you can apply after this session.
Dreamforce passes are going fast. Don’t miss the most valuable event in business. Register now.
Register to join Dreamforce in San Francisco, or stream it free on Salesforce+.
Dreamforce is almost here and passes are going fast. Don’t miss your last chance to save $400 by September 26.
Sessions
Explore Dreamforce 2025 sessions.
Browse 1,600+ sessions from 2025 by role, topic, or industry and get inspired for the innovation coming to Dreamforce 2026.
Architecting Agentforce: The 5 Design Pillars to AI Success
Required equipment: Salesforce Events Mobile App required. Please bring your headphones and use Live Session Audio to hear audio and view captions.
, Specialist Master, Deloitte
Role: Architect
Product: Agentforce
Topic: Agentic AI
Session Level: Advanced
Session Type: Theater
Session Summary: This session, titled 'Architecting Engineers: The Five Design Pillars to AI Success,' provided a comprehensive guide to designing scalable, secure, and effective AI solutions using Salesforce tools and platforms. The speaker emphasized the importance of aligning AI proof-of-concepts (POCs) with robust architectural principles to ensure they evolve into production-ready, trusted, and valuable products. The five design pillars covered were data quality, security, orchestration, performance, and adoption, with each pillar addressing critical factors such as data governance, secure access, scalable orchestration through tools like MuleSoft, performance optimization, and user adoption strategies. Attendees were equipped with actionable deliverables, including data models, security flow diagrams, integration patterns, and performance metrics, ensuring they can create AI solutions that deliver measurable business value and meet enterprise-grade standards for scalability and resilience.
Key Takeaways:
• The five design pillars—data quality, security, orchestration, performance, and adoption—are essential for transforming AI POCs into scalable and valuable enterprise solutions.
• Deliverables such as canonical data models, security flow diagrams, and integration patterns must be defined during the design phase to ensure seamless scalability and reliability of AI agents.
• Engineering for performance and ensuring robust orchestration through Salesforce tools like MuleSoft are critical for building efficient, cost-effective, and resilient AI solutions that meet enterprise demands.
Session Summary Feedback: Report a Concern
Available Now on Salesforce+: Watch now
