#Agentforce322 diskutieren mit

Hi @Pavan Kumar Gandi

, 

Agentforce on its own isn’t what’s changing (or rather challenging) the developer role. The whole stack is moving together. Data Cloud gives agents unified, real-time context. Model choice is opening up, with partner models like Claude now usable inside Agentforce. And the trust layer, Flow and Apex sit underneath all of it. Agentforce is just the visible front end. 

So the developer role is shifting from building screens and triggers to building the pieces agents run on, including better grounded data, efficent decision control mechanishm, tighted security controls, and more. According to me, it is time to leave the 'Developer' tag, and start looking at the broader perspective, where you would be expected to anything and everything (before AI replaces your dev-only skills). 

Hi @Jonathan Stoop You already have 3 years of Salesforce Admin experience, which is a good foundation for learning AI and Agentforce.

My suggestion is to focus more on hands-on implementation rather than only completing Trailhead modules.

Here is a simple learning roadmap:

1. AI Basics – Understand LLMs, prompts, RAG, and AI agents.
2. Agentforce – Learn Agent Builder, subagents, and actions.
3. Flow & Apex – Start with Flow actions, then learn basic Apex.
4. Data 360 – Understand data grounding and knowledge retrieval.
5. APIs & Integration – Learn how AI agents connect with external systems.
6. Real Projects – Build an Employee Support Agent or Customer Service Agent.
7. Testing – Practice testing agent responses, actions, and security.

Useful Learning Resources:

1. Agentblazer Learning Journey: 
   https://trailhead.salesforce.com/agentblazer
2. Agentforce Developer Guide: 
   https://developer.salesforce.com/docs/ai/agentforce/guide/get-started.html
3. Agentforce Project Examples: 
   https://developer.salesforce.com/docs/ai/agentforce/guide/agent-examples.html
4. Custom Agent Actions: 
   https://developer.salesforce.com/docs/ai/agentforce/guide/get-started-actions.html
5. Agentforce APIs and SDKs: 
   https://developer.salesforce.com/docs/ai/agentforce/guide/get-started-agents.html
6. Agentforce Trailblazer Community: 
   https://trailhead.salesforce.com/trailblazer-community/topics/agentforce
7. Salesforce Data 360 Overview: 
   https://sfdcindia.com/blog/salesforce-data-360

My View:

Try to build 2–3 real-time projects, document your work, and share your learning on Trailblazer or LinkedIn.

You don't need to master everything at once. Start with one simple use case and gradually move to advanced implementations.

Practical project experience will help you build confidence and demonstrate your skills to future employers.

Hope this helps!

Hi @Pavan Kumar Gandi I believe Salesforce testers will still be important in the AI era. AI can reduce repetitive testing, but human testers are needed to validate business scenarios, accuracy, and security.

Skills Salesforce testers should focus on:

1. Automation Testing – Selenium, Playwright, and API testing.
2. Agentforce Testing – Validate agent responses, subagents, and actions.
3. AI Testing – Check response accuracy and unexpected behavior.
4. Salesforce Testing – Flows, Apex, LWC, and integrations.
5. Security Testing – Validate permissions and data access.
6. Problem-Solving – Understand business requirements and identify issues.

Real-Time Example:

A customer asks Agentforce to update an order. Testers should verify whether the correct action runs, the record is updated, and unauthorized users are restricted.

My View:

Manual and automation testing will continue to be important. Testers who learn Agentforce and AI testing will have better opportunities.

Useful Salesforce References:

1. Agentforce Testing Center: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing
2. AI-Generated Test Cases: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing/use-ai-to-generate-tests
3. Agentforce Testing Criteria: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing/set-up-testing-criteria

Similar Trailblazer Discussion:

1. How Do You Test Agentforce Properly? 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000jAUii0AG

Don't depend only on manual testing or automation tools. Start learning Agentforce Testing Center and AI testing concepts to stay relevant.

Hi @Pavan Kumar Gandi I believe Salesforce Admins can get a lot of value from Agentforce, especially when they already have strong Flow and configuration skills.

How Admins can use Agentforce:

1. Record Management – Create, update, and retrieve records using Flow actions.
2. Business Automation – Automate approvals, follow-ups, and repetitive tasks.
3. User Support – Help employees find information and answer common questions.
4. Case Management – Summarize cases and assist support teams.
5. Setup Assistance – Use Setup with Agentforce for supported admin tasks and troubleshooting.
6. Custom Agents – Build agents using Flow, Prompt Builder, and standard actions.

Real-Time Example:

An employee asks Agentforce to create a support request. The agent collects the required details, calls a Flow to create a Case, and shares the Case number.

My View:

Yes, Admins can build useful Agentforce solutions without much Apex knowledge. Apex becomes important when requirements involve complex logic or custom integrations.

Useful Salesforce References:

1. Agentforce Customization with Flows: 
   https://trailhead.salesforce.com/content/learn/modules/agent-customization-with-flows
2. Create Agent Actions Using Flow: 
   https://trailhead.salesforce.com/content/learn/modules/agent-customization-with-flows/add-a-flow-as-an-agent-action
3. Agentforce Customization Overview: 
   https://trailhead.salesforce.com/content/learn/modules/agent-customization-quick-look

Similar Trailblazer Discussions:

1. How Are You Using AI to Improve Salesforce Workflows? 
   https://trailhead.salesforce.com/de/trailblazer-community/feed/0D5KX00000jARpK0AW
2. Can Setup with Agentforce Be Used for Admin Tasks? 
   https://trailhead.salesforce.com/it/trailblazer-community/feed/0D5KX00000pNL8j0AG

Admins should strengthen their Flow skills and start learning Agentforce. No need to become an Apex expert before getting started.

Hi @Pavan Kumar Gandi Good question! I believe Agentforce can help Salesforce QA teams reduce repetitive work, but it cannot completely replace traditional testing tools or human testers.

How QA teams can use Agentforce:

1. Test Case Creation – Generate test scenarios from business requirements.
2. Test Data Preparation – Create sample records using Flow or Apex actions.
3. Defect Analysis – Analyze error messages, logs, and possible root causes.
4. Regression Testing – Help identify and organize test cases after changes.
5. Agentforce Testing Center – Validate agent responses, subagent selection, and actions.

Real-Time Example:

A QA team can build a custom agent to generate test cases for an Opportunity approval process, prepare sample data, and summarize test failures.

My View:

Agentforce can support QA activities, but Selenium, Playwright, Apex tests, and manual testing are still important for complete test coverage.

Useful Salesforce References:

1. Agentforce Testing Center: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing
2. Testing Criteria: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing/set-up-testing-criteria
3. Agentforce Testing Methods: 
   https://trailhead.salesforce.com/content/learn/modules/agentforce-agent-testing/trust-your-agents

Similar Trailblazer Discussions:

1. How Do You Test Agentforce Before Production? 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000jAUii0AG
2. Agentforce & AI-Powered Testing: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000Tj0RX0AZ
3. Agentforce for Regression Testing: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000mmbsT0AQ

Use Agentforce to assist testers with repetitive activities, but continue using traditional automation tools and human validation for critical business scenarios.

Hi @Pavan Kumar Gandi Good question! I believe Data 360 plays an important role in making Agentforce more useful, especially when customer data is stored across multiple systems.

How Data 360 helps Agentforce:

1. Unified Customer View – Combines customer information from different systems.
2. Identity Resolution – Helps identify the same customer across multiple data sources.
3. Better Responses – Provides relevant customer context to help Agentforce answer more accurately.
4. Unstructured Data – Makes information from documents and knowledge articles searchable through RAG.
5. Personalization – Helps agents provide responses based on customer history and activities.

When is CRM data enough?

If an agent only needs Accounts, Contacts, Opportunities, or Cases already available in Salesforce, CRM data with the right agent actions may be enough.

When do we need Data 360?

When customer information is spread across Salesforce, external databases, websites, and other systems, or when we need unified profiles and advanced knowledge retrieval.

Real-Time Example:

A customer asks, "Where is my order, and can I return it?"

• Salesforce CRM provides customer and case details.
• An external order system provides order and delivery status.
• Data 360 can bring relevant information together.
• Agentforce uses that information and configured actions to answer the customer or initiate a return.

My View:

Data 360 is not mandatory for every Agentforce use case. Start with CRM data if it meets the requirement. Use Data 360 when there is a clear need for unified data, identity resolution, or advanced retrieval.

Also, remember that Data 360 alone doesn't guarantee accurate responses. Data quality, permissions, and proper agent configuration are equally important.

Useful References:

1. Salesforce Data 360 Learning: 
   https://trailhead.salesforce.com/en/search?keywords=data%20cloud
2. Agentblazer Learning Journey: 
   https://trailhead.salesforce.com/agentblazer
3. Data 360 Overview – SFDCINDIA: 
   https://sfdcindia.com/blog/salesforce-data-360

Similar Trailblazer Discussions:

1. CRM Grounding vs. Data Cloud for Agentforce: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000mn6140AA
2. Data Cloud with Agentforce for Complex Cross-Object Queries: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000Tm4vA0AR
3. Data Cloud and AI Agent Dependency: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000Tiycu0AB
4. Account Records and Attached Files for Agentforce: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000KCual0AD

Agentforce performs the work, and Data 360 helps provide the right customer context to make better decisions.

Hi @Pavan Kumar Gandi Good question!

I believe Agentforce will definitely change how Salesforce project teams work, but it won't completely replace existing roles.

Some repetitive tasks may reduce, while new AI-related responsibilities and opportunities will increase. 

Here is how I see the roles evolving:

1. Admins – Manage AI agents, Flow automation, permissions, and security.
2. Developers – Build custom actions, Apex, APIs, and integrations.
3. Testers – Test agent responses, accuracy, security, and unexpected behavior.
4. Business Analysts – Identify business processes where AI can add value.
5. Architects – Design secure and scalable AI solutions.

We may also see new roles like Agentforce Specialists, AI Architects, and AI Testing Engineers. 

My View:

Over the next 3–5 years, some project teams may become smaller for routine work, but complex AI projects will still need skilled professionals.

I believe Admins, Developers, Testers, and Architects who learn Agentforce and Data 360 will have better opportunities. 

Useful Salesforce References:

1. Agentblazer Learning Journey: 
   https://trailhead.salesforce.com/agentblazer
2. Agentforce Trailblazer Community: 
   https://trailhead.salesforce.com/trailblazer-community/topics/agentforce

Related Trailblazer Discussions:

1. Learning AI and Implementation: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000ao2UO0AY
2. Upskilling from Testing to Agentic AI: 
   https://trailhead.salesforce.com/trailblazer-community/feed/0D5KX00000mjgHW0AY

Don't focus only on whether AI will reduce jobs. Focus on how we can use AI to improve our existing skills and deliver better Salesforce solutions.

Hi @Pavan Kumar Gandi I believe Salesforce Admin roles will still be needed in the AI era, but the way Admins work will definitely change.

AI can handle repetitive tasks, but companies still need Admins to understand business requirements, manage security, maintain data quality, and solve real-time issues.

To stay relevant, Admins should focus on:

1. Flow – Business process automation.
2. Agentforce – AI agents and actions.
3. Data 360 – Data management and integration.
4. Security – Permissions, sharing, and data access.
5. Integration – APIs and connected systems.

In my view, AI won't replace Salesforce Admins completely, but Admins who learn AI will have better opportunities.

The key is to keep learning and move beyond traditional Admin skills.