In this episode, we welcome back Mohith Shrivastava to delve into the groundbreaking features of the Salesforce Winter '25 release. We discuss Agentforce, Salesforce's innovative AI agentic solution that leverages tools like Flows, Apex, and Prompt Builder to autonomously execute tasks. The episode also highlights advancements in Agent Builder (fka Copilot Builder) for creating assistive and autonomous agents, and the introduction of Agentforce for Developers.

Further, we explore significant updates in Salesforce Data Cloud, including Retrieval-Augmented Generation (RAG) capabilities, vector search, and the much-anticipated beta release of Data Cloud sandboxes. Finally, we touch on the new local development feature for Lightning Web Components (LWC) that provide real-time previews and rapid iterations.

Tune in to learn how these Winter '25 updates can elevate your Salesforce development and data management capabilities.

Show Highlights:

• Enhancements to Copilot Builder with new topic-based action categorization for assistive and autonomous agents.
• A chat agent for Salesforce developers to explain code, write test cases, and generate documentation.
• Upcoming features like Data Cloud sandboxes for secure testing and new dependencies between Data Cloud and Salesforce platform packages for ISVs.
• Lightning Web Components (LWC) advancements, including local development for real-time previews and rapid iterations.
• New Apex feature preventing collection modification during iteration to enhance code stability and consistency.

Links:

• The Salesforce Developer’s Guide to the Winter ’25 Release - https://developer.salesforce.com/blogs/2024/09/winter25-developers

Alba Rivas, a passionate Salesforce Developer Advocate, shares her insights on the power of Salesforce's Models API. Discover how this API enables developers to seamlessly integrate large language models via REST API or Apex, eliminating the need for custom code or prompt templates. Alba highlights the critical role of the Einstein Trust Layer in ensuring data security, compliance, and efficient integration with external models. Gain knowledge on AI implementation, including maintaining audit trails, monitoring data for toxicity, and using human feedback for refining AI systems.

We also explore the flexibility and capabilities of the Models API, offering a centralized interface for developers, along with resources like Postman collections, codeLive sessions, and an upcoming Trailhead module to support your journey in mastering generative AI with Salesforce.

Show Highlights:

• Discussion on the Einstein Trust Layer's role in ensuring security, data masking, and compliance for generative AI applications.
• Comparison between using prompt templates and the Models API.
• Importance of maintaining an audit trail.
• Focus on the value of human feedback in refining AI systems and the benefits of a centralized interface for developers.
• Insights into practical AI implementation and the flexibility offered by the Models API.

Links:

• Postman Collection - https://www.postman.com/salesforce-developers/workspace~34382471-0c97-40e5-a206-f947271665c4/collection/onih7sc/models-apis-beta
• Einstein Models API Documentation - https://developer.salesforce.com/docs/einstein/genai/guide/models-api.html
• codeLive: Exploring the New Models API (Beta) - https://www.youtube.com/watch?v=hf95_ghGeNI

In this episode, we sit down with expert Akshata Sawant to discuss MuleSoft Intelligent Document Processing (IDP). Akshata delves into how IDP uses AI to extract and organize data from a variety of unstructured documents, such as purchase orders, invoices, and driving licenses, using AWS Textract for exact information extraction. The discussion highlights IDP's seamless integration with multiple systems via REST API endpoints and its application in automating the processing of daily invoices received via email.

Learn about IDP's ability to handle different document structures without needing multiple templates and the crucial role of human oversight when the AI's confidence ratio falls below 80%, ensuring accuracy and reliability. This episode offers a comprehensive look at how IDP can revolutionize document workflows, balancing automation with human intervention, and provides valuable resources for further learning.

Show Highlights:

• Using AI to extract and organize data from unstructured documents
• Integration of AWS Textract for exact information extraction and seamless integration via REST API endpoints
• Accessing and using IDP through the Anypoint Platform with ease
• Managing different document structures without multiple templates and the role of human oversight for accuracy

Links:

• Getting Started with Intelligent Document Processing: Part 1 - https://www.youtube.com/watch?v=Yi160vy7gg4
• Getting Started with Intelligent Document Processing: Part 2 - https://www.youtube.com/watch?v=iGvAypuzAvw
• How to Get Started With Intelligent Document Processing - https://blogs.mulesoft.com/dev-guides/how-to-get-started-with-intelligent-document-processing/