Catalogue Review
A catalogue review workflow that turns inconsistent product information into clear, reviewable suggestions.
Inside the workflowThe AI studio / By Gokul Raj Arimbra
Less repetition.
More possibility.
Small, purposeful agents and tools for the work between the big ideas. Designed around useful inputs, visible progress and a person in control.
A practical point of view
Start with the work.
Build the right tool.
My background is hands-on commerce: product catalogues, storefronts, campaigns and the systems behind them. That is the lens I bring to AI.
The 8 studies below explore focused tasks across commerce, customer communication, research and marketing. Each makes a proposal visible and keeps a person in control. These are interactive prototypes with fictional sample data.
Selected experiments
Useful tasks. Clearer workflows.
Research Brief
An interactive research workflow for organising a brief, finding recurring themes and preparing a reviewable outline.
Inside the workflowContent Review
A focused content review workflow for clearer interface labels, useful error messages and meaningful confirmations.
Inside the workflowProduct branch agent
An interactive study for reviewing product families, variants and category branches before changing a catalogue.
Inside the workflowCustomer communication agent
A customer-reply workflow that identifies missing context and prepares a draft for human review.
Inside the workflowGoogle Ads agent
An interactive Google Ads study for checking campaign structure, search intent and landing-page relevance.
Inside the workflowMeta Ads agent
A Meta Ads workflow study for reviewing audience intent, creative direction and pre-publication checks.
Inside the workflowLinkedIn post agent
An interactive writing workflow that shapes project notes into a LinkedIn draft with a separate approval step.
Inside the workflow The agent studioInteractive simulation · Sample data
See the system
think in steps.
Choose a workflow, inspect the sample input and review the proposed output. You make the final call.
01 / The task
Find inconsistencies in a small sample product catalogue.
Sample input
Product: Wireless headphones · Category: Audio · Compatibility: missing · Related products use the Headphones category.
- 01Inspect sample product fields
- 02Compare naming and categories
- 03Prepare suggested corrections
02 / The proposed outputReady
Your review workspace.
Run the example to reveal the proposed changes.
This browser simulation demonstrates the workflow and interface. Its sample responses are predefined; it does not call an AI model, access a store or publish changes.
Workflow designHuman reviewTypeScriptReact
The AI landscape
Different tools.
A considered approach.
A selection of AI platforms and workflow tools. The task comes first; the tool follows.
Claude
ChatGPT
Gemini
DeepSeek
Perplexity
Microsoft Copilot
GitHub Copilot
n8n
The design principles
Trust is part of the interface.
Give the task a shape
Start with a defined input, a specific job and an output someone can actually use. Small boundaries make a workflow easier to understand.
Make progress visible
Show where the work is, what has been prepared and what needs attention. Useful status is more valuable than a mysterious loading spinner.
Keep a person in control
Make suggestions reviewable. Separate proposed changes from accepted ones, with a clear moment to check the details.
Where this can help
Build around the friction
in your business.
- Product information and catalogue operations
- Research organisation and content preparation
- Internal tools and approval workflows
- API-connected commerce systems