“The only thing we’re going to do at Salesforce is Agentforce. » This is what Marc Benioff, CEO of Salesforce, said in September during the presentation of AI agents. This week, the company showed what that means by unveiling a significantly improved Agentforce 2.0.
Among its new features, four stand out: the library of predefined skills and workflow integrations for rapid customization, the ability to deploy Agentforce in Slack, improvements to Atlas Reasoning Engine and Agentforce Testing Center. Here’s an overview of these and how they can be used.
Skills and Integrations Library
This library covers key business applications such as CRM, Slack, and Tableau, as well as AppExchange partner integrations. These out-of-the-box skills simplify the process of creating bespoke AI agents. This eliminates complex and time-consuming integration issues. The library contains predefined Slack actions such as Create Canvas and Message Channel that allow teams to create agents for their Slack workflows.
Here are some examples:
CRM skills allow you to create agents to automate and improve customer experiences. These agents can perform tasks such as lead nurturing, sales coaching, scheduling service engagements, among other functions. For example, the Sales Development skill can create agents who nurture leads based on predefined rules. The Sales Coaching skill allows agents to actively participate in sales calls, providing real-time feedback to reps and improving their performance. MuleSoft integrations enable businesses to use Agentforce across their entire business ecosystem. For example, it can connect Agentforce to its ERP system to automate order fulfillment processes. It can also integrate with marketing automation platforms to trigger personalized email campaigns based on customer interactions. AppExchange enhances Agentforce’s capabilities by giving users access to specialized skills created by partners.
Agentforce in Slack
With this, Slack becomes the central point for employees to interact with Agentforce. They can access information, trigger workflows, and manage tasks directly in Slack. This means they don’t need to switch between apps.
Agentforce can also automate tasks in Slack. For example, Accenture uses it to provide employees with quick updates on important accounts. It helps automate document retrieval and provides recommendations to resolve business issues. It also streamlines proposal creation and viewing and improves employee collaboration, including providing real-time updates, retrieving documents, providing proactive recommendations, and creating proposals.
Dig Deeper: SharkNinja embarks on its Salesforce AI adventure
Agentforce also uses Slack Enterprise Search. It learns from conversations on Slack channels and direct messages. This allows agents to provide more relevant responses and actions based on the context of conversations.
Improved Atlas reasoning engine
The Atlas reasoning engine has been enhanced with advanced reasoning and augmented retrieval generation (RAG) so agents can process more complex, multi-step questions and provide more accurate answers.
For example, RBC uses Atlas and Data Cloud with Agentforce to support its financial advisors. Data Cloud integrates all client data into a unified system, giving advisors a complete view of their clients. Atlas can answer complex customer questions. To do this, it uses multi-step reasoning. It also leverages RBC data and business logic. This helps provide a reliable response.
There is an old Russian saying: “Trust, but verify.” Enhanced RAG features allow users to do this by providing in-line quotes with agent responses. Citing the exact sources used builds confidence in the AI results. It also helps users check for accuracy.
Agentforce Test Center
The testing center allows companies to test their AI agents before they are used. This ensures that agents follow instructions, remain factual, and work quickly. This helps mitigate risk and ensure agents operate consistently and reliably. It also provides tools that make it easier to test, audit, and scale agents.
Practicing and Preaching
Salesforce uses the technology internally. It says help.salesforce.com agents now handle 83% of customer support requests independently, with human escalations decreasing by 50% since implementation two weeks ago.
“Suddenly, as a CEO, I’m not just managing human beings, but I’m also managing agents,” Benioff said. “Today there is a real agentic layer around the platform. This is not a fanciful vision of the idea of the future, but of what is happening right now. »
