Jayanth Thathapudi

I Tested Claude Code Agent View with a Simple Salesforce Workflow

Jayanth Thathapudi — Mon, 15 Jun 2026 02:21:19 GMT

I tested the new Claude Code Agent View in the terminal with a simple Salesforce workflow.

The part that stood out was not that it can run more AI tasks.

It was that I could finally see the work.

Before Agent View, my Salesforce workflow could get messy fast.

If I was working across multiple Salesforce projects, I would open multiple terminal tabs:

- One for checking metadata
- One for the validation rule
- One for the trigger
- One for test notes
- One for release cleanup

That works for a while.

Then it becomes hectic.

You start asking basic questions:

- Which tab is still running?
- Which task already finished?
- Which one needs my input?
- Which Salesforce project was this session for?
- Did I already review that output?

Agent View gave me one terminal screen where I could dispatch, monitor, peek, attach, and step away without losing the thread.

For the test, I kept the Salesforce demo simple:

- An Opportunity amount validation rule
- An Account rating trigger

What clicked for me was the dashboard.

When I came back, I did not have to remember what was happening. The rows told me. One session was still working. One was waiting on me.

That “waiting on me” state mattered because, in my old setup, it was invisible. It lived in another tab, behind a scroll, in a conversation I had half-forgotten.

That is where Agent View felt useful.

Not “let AI build the whole project.”

More like:

Give each small Salesforce task a clear session, then supervise the outputs.

I wrote the full walkthrough here:
https://www.jthathapudi.com/blog/i-tested-claude-code-agent-view-for-salesforce-heres-what-i-learned

API vs MCP: Different Layer, Different Purpose

Jayanth Thathapudi — Sun, 07 Jun 2026 23:31:44 GMT

API vs MCP is not about which one is better.

It is about what each one is designed for.

APIs have been one of the foundations of modern software integration. They provide defined interfaces for software systems to communicate with each other.

A traditional application usually knows exactly which API to call, what request to send, and what response to expect.

MCP — Model Context Protocol — is different.

MCP provides a standardized way for AI applications to access tools, resources, prompts, and context from external systems.

That distinction matters because AI applications often work differently from traditional applications.

A traditional app usually follows a predefined flow.

An AI application may need to:
- discover available tools
- retrieve relevant context
- choose which action to take
- call the right tool
- synthesize a response for the user

This does not mean MCP replaces APIs.

In many real-world architectures, MCP can sit on top of existing APIs, databases, files, and workflows.

A simple mental model:

APIs expose functionality.

MCP makes functionality, tools, and context accessible to AI applications.

APIs are useful when the operation and integration flow are known.

MCP is useful when an AI application needs standardized access to tools and context across systems.

Different purpose.
Different layer.
Often used together.

Better together: API + MCP.

As AI applications become more common in enterprise systems, this distinction will become increasingly important.

The question is not “API or MCP?”

The better question is:

“How do we expose existing systems in a way that both applications and AI agents can safely use?

I found a use case for Claude Cowork I didn't expect

Jayanth Thathapudi — Tue, 17 Feb 2026 21:29:11 GMT

Hey there,

I found a use case for Claude Cowork I genuinely didn’t expect.

I was working through Salesforce Translation Workbench — the usual painful workflow. Export STF files. Clean them up line by line, removing managed package items and unsupported elements. Create an Excel for the business team. Get translations back. Build the import file. Import. Hit errors. Fix them. Re-import.

Repeat for every version, every environment.

If you’ve done this, you know. It’s hours of careful, tedious file work.

I decided to throw the whole thing at Claude Cowork — Anthropic’s desktop agent for non-developers. You give it access to a folder, upload files, and work through tasks conversationally.

What Actually Happened

I walked through five steps with Cowork:

STF cleanup — It identified and removed managed package items and unsupported TextTemplate keys. Hours of line-by-line review became minutes.
Audit reports — It generated structured Excel files documenting every change and why. Ready for the person doing the load.
Import file creation — It built the import STF from a translated Excel with the correct header and formatting.
Error log parsing — I uploaded Salesforce error logs, and it came back with a fixed file each time. Character limit errors, duplicate keys, translation type mismatches — it handled each one differently.

The error-fix-reimport cycle — the part that normally takes an hour per iteration — became a few minutes of conversation.

The Honest Take

It’s not perfect. You still need to review files before importing. Some Salesforce-specific issues still need human escalation.

But every single step got faster. And the conversational loop — upload error log, get fixed file, re-import — was surprisingly effective.

Try It Yourself

The full walkthrough is on the blog with real error messages, file formats, and the honest limitations: I Tested Claude Cowork on the Salesforce Translation Workbench Workflow (It Handled the Whole Thing)

Until next time,
Jay

Step-by-Step: Create Your Own Salesforce AI Assistants using Claude Code & Salesforce DX MCP

Jayanth Thathapudi — Mon, 15 Sep 2025 14:36:34 GMT

Watch my ApexHours session on 'Step-by-Step: Create Your Own Salesforce AI Assistants using Claude Code & Salesforce DX MCP'. Thanks to Amit Chaudhary and the ApexHours team for having me!

I demonstrated two working AI agents that deploy directly to your Salesforce org:

🔹 Admin Agent: Creates validation rules through natural language - built and deployed

🔹 Developer Agent: Writes complete Apex with test classes - generated trigger with handler pattern and test coverage

Created and tested both the validation rule and the apex trigger in the session. These agents handle the repetitive coding tasks, letting you focus on solving business problems.

The companion blog post has all the agent prompts ready to copy-paste. Works with any IDE - VS Code, IntelliJ, Cursor, Windsurf, Cline, & Others.

What Salesforce tasks would you automate first?

ApexHours:

#Salesforce #ApexHours #AI #SalesforceAdmin #SalesforceDeveloper #ClaudeCode #SalesforceDXMCP #ValidationRules #ApexTriggers #AIAgents #SalesforceAutomation #TestClasses #SalesforceAI #DevOps #SalesforceProductivity #AIAssistants #SalesforceBestPractices #DeveloperTools

Build Your Own Salesforce Developer AI Agent with Claude Code

Jayanth Thathapudi — Thu, 21 Aug 2025 14:39:12 GMT

The more I explore Salesforce DX MCP with Claude Code, the more impressed I am with what's possible.

Episode 03 of my Salesforce AI Agent series is now live!

I built an AI assistant for developers that writes clean Apex triggers, creates Lightning Web Components, and generates test classes with 95%+ coverage. Using Claude Code with Salesforce DX MCP, this agent understands Salesforce development patterns and best practices.

What I Tested

This isn't theoretical - I tested it on real development tasks:

Built Account triggers with proper handler patterns
Created Lightning components with Apex controllers
Generated comprehensive test classes covering edge cases
All following Salesforce coding standards automatically

This isn't about replacing developers - AI generates the code and components faster while you maintain full control to review, refine, and deploy. Focus on architecture and business logic, let AI handle the repetitive coding work.

No more syntax errors, missing bulkification, or forgetting test coverage requirements.

Resources

YouTube

Blog Post: https://www.jthathapudi.com/blog/build-your-own-salesforce-developer-ai-agent-with-claude-code

GitHub: https://github.com/jthathapudi/sf-ai-agents/blob/master/.claude/agents/salesforce-dev-pro.md

The Series Continues

This is Episode 3 of my Salesforce AI Agent series. Each episode focuses on a different role:

- Episode 2: Admin AI Agent

- Episode 3: Developer AI Agent (this one!)

- More roles coming soon...

Your Thoughts?

What type of Salesforce AI agent would you be most excited to see next?

Want to know what permissions & groups a Salesforce user has?

Jayanth Thathapudi — Wed, 20 Aug 2025 03:55:53 GMT

Try this:

Go to Setup → Users → Select any user → Click "View Summary"

You'll see everything in one place:

- User permissions

- Object & field permissions

- Custom permissions

- Permission sets & groups

- Tabs they can view

- Public groups & queues

No more searching through layers of settings. Just one click.

Youtube: Watch Here

#Salesforce #Tips #CRM #Productivity

Unlock Salesforce Data Cloud: From Raw Data to Customer 360

Jayanth Thathapudi — Wed, 20 Aug 2025 03:51:26 GMT

Learn the data transformation journey with the three foundational building blocks that turn raw data into unified customer profiles:

✅ DSO (Data Source Object) – Your raw data staging area. Captures incoming data from any source in its original format.

✅ DLO (Data Lake Object) – Your transformation layer. Cleanses, structures, and optimizes data for long-term storage and analysis.

✅ DMO (Data Model Object) – Your unified business view. Creates dynamic, real-time customer profiles by combining data from multiple DLOs.

Learn more about DSO, DLO, DMO: https://www.jthathapudi.com/blog/from-raw-to-ready-understanding-dsos-dlos-and-dmos-in-salesforce-data-cloud

#SalesforceDataCloud #CustomerDataPlatform #DataUnification #Customer360 #SalesforceArchitecture #DataManagement #RealTimeData #DataDrivenBusiness #PersonalizedCustomerExperience

Unlock the Power of Salesforce Data Cloud

Jayanth Thathapudi — Wed, 20 Aug 2025 03:43:12 GMT

Discover how Salesforce Data Cloud simplifies data management with its five core capabilities:

✅ Connect – Unify data from apps, devices, and CRMs.

✅ Harmonize – Organize data into a standard structure.

✅ Unify – Create a 360-degree customer view.

✅ Analyze & Predict – Gain actionable insights with Tableau or other AI-powered tools.

✅ Activate – Deliver real-time personalized experiences.

Learn more about Data Cloud: https://www.jthathapudi.com/blog/salesforce-data-cloud-understanding-its-power-use-cases-and-provisioning

#SalesforceDataCloud #CustomerDataPlatform #Agentforce #SalesforceAI #RealTimeData #Customer360 #SalesforceInnovation #DataDrivenBusiness #PersonalizedCustomerExperience #TrustedAI