AI-SDLC Experiences: AI-SDLC Mob Programming in Action

Welcome to AI-SDLC “Mob Programming”

I recently created a repository called: AI-SDLC-Mob-Programming-Testing

The goal is to create a reusable collection of AI-SDLC playbooks that help teams collaborate more effectively and produce implementation-ready outcomes faster.

Repository: https://github.com/javakishore-veleti/AI-SDLC-Mob-Programming-Testing

The purpose of this repository is to provide reusable AI-SDLC playbooks that help teams move from requirements discussions to implementation-ready outcomes. The repository is built around a simple idea: AI-SDLC Mob Programming

What Is AI-SDLC Mob Programming?

AI-SDLC Mob Programming is a collaborative software delivery approach where Product Owners, Architects, Developers, QA Engineers, DevOps Engineers, Security Engineers, and AI work together during design and planning activities.

The objective is not to generate code.

The objective is to create implementation readiness.

A successful session should answer questions such as:

  • What tools are required?

  • What access is required?

  • What APIs are needed?

  • What architecture is required?

  • What security concerns exist?

  • What edge cases exist?

  • What testing strategy is required?

  • What implementation artifacts should be produced?

Why SKILL.md Playbooks?

Many teams already use AI.

However, AI often lacks structure.

The repository introduces reusable SKILL.md playbooks that guide collaborative sessions.

Each playbook defines:

  • Session purpose

  • Required tools

  • Required logins

  • Required system scopes

  • Ideal participant roles

  • Session agenda

  • Expected outcomes

  • Deliverables

The playbook becomes a facilitation framework.

Shopify Middleware Discount Segmentation Example

One of the first playbooks focuses on:

Guest Customers
Logged-In Customers
VIP Customers
Wholesale Customers

A typical team must answer several questions:

Business Questions

  • Which customer segments exist?

  • Which discounts apply to each segment?

  • Can discounts stack?

  • How should guest customers be handled?

Architecture Questions

  • Where is customer segmentation performed?

  • Which middleware services are required?

  • What Shopify APIs are needed?

  • How should caching work?

API Questions

Example:

POST /discount/evaluate

The team defines:

  • Request model

  • Response model

  • Authentication

  • Error handling

  • Observability requirements

Testing Questions

The playbook helps generate scenarios for:

  • Guest customers

  • Logged-in customers

  • VIP customers

  • Wholesale customers

  • Discount conflicts

  • Expired discounts

  • Shopify API failures

  • Checkout validation failures

How This Differs From Traditional Backlog Refinement

Traditional backlog refinement typically focuses on preparing stories for future implementation.

AI-SDLC Mob Programming focuses on implementation readiness.

Instead of ending with:

User Story
Acceptance Criteria
Story Points

the session can produce:

Architecture Decisions
API Contracts
Business Rules
Test Scenarios
Edge Cases
Implementation Checklists
Backlog-Ready Stories

The outcome is a deeper level of shared understanding across the team.

Current Playbooks

The repository currently contains:

  • Discount Segmentation

  • Checkout Customization

  • Inventory Synchronization

  • OMS / ERP Integration

Each playbook follows the same AI-SDLC Mob Programming approach.

Looking Ahead

Although the initial examples use Shopify Headless Middleware, the framework is intentionally broader.

Future playbooks can support:

  • Enterprise Integrations

  • Cloud-Native Architectures

  • Microservices

  • Data Platforms

  • AI Agents

  • Quality Engineering

  • DevOps Workflows

The goal is to create a reusable collection of AI-SDLC playbooks that help teams collaborate more effectively and produce implementation-ready outcomes faster.

Repository: https://github.com/javakishore-veleti/AI-SDLC-Mob-Programming-Testing