AI-Assisted Software Architecture & Domain Discovery

Indu Alagarsamy 2-Day Workshop
Mon-Tue, Sept 21-22

Learn how to use AI tools as a force multiplier for architectural and domain discovery in complex existing systems.

Your company has gone all-in on AI. You've got Copilot, Claude Code, Cursor, maybe even all of them. Leadership expects you to move faster. But the hard problems haven't changed: sprawling codebases where no one fully understands how things interact, domain concepts buried implicitly across thousands of files, and knowledge that lives in a few people's heads instead of in documentation.

AI won't simplify a complex system. But it can dramatically accelerate your understanding. And understanding is what unlocks every decision that follows. This workshop teaches you how.

What you will learn

Over two full days, you'll learn how to use AI tools as a force multiplier for architectural and domain discovery:

Codebase exploration: Rapidly build a mental model of an unfamiliar or poorly documented system, its components, responsibilities, and interactions
Domain concept extraction: Surface the implicit domain model hiding in your codebase: entities, relationships, domain events, communication patterns, and business rules that no one has documented
Bug pattern analysis: Determine whether a defect is a one-off symptom or a systemic issue embedded across your bounded context
Dependency mapping and coupling analysis: Trace the dependencies that determine what's actually hard to change. Find the load-bearing components, shared databases, and hidden coupling
Generating architectural documentation from code: Produce living documentation such as sequence diagrams, interaction maps, and data flow descriptions directly from your codebase
Knowing where AI can't help: Systems are built for people, and people don't always decide on the merits. We'll combine structured trade-off evaluation techniques like Double Diamond with facilitation practices that help stakeholders reach genuine shared understanding

While our demos will use Claude Code, our goal as a group is to learn from one another. If you have a different tool installed, bring it. Different tools often surface different insights on the same problem.

Who this workshop is for

Software Architects and Principal Engineers who want to add AI-assisted techniques to their architectural analysis toolkit
Senior Developers and Tech Leads responsible for understanding, evolving, or modernizing existing systems
Consultants and Coaches who help organizations understand and improve their software systems
Anyone who regularly needs to make sense of complex existing codebases, whether to add features, plan migrations, onboard to a new team, or assess technical debt

This workshop is designed for people who work with already existing systems. If you only work on greenfield projects with simple domains, most of what we cover won't resonate yet.

Agenda

Day 1

Codebase exploration and building mental models
Domain concept extraction from existing code
Bug pattern analysis across bounded contexts

Day 2

Dependency mapping and coupling analysis
Generating architectural documentation from code
Where AI can't help: facilitation and stakeholder alignment

How it works

You'll work with a realistic sample codebase, alternating between demos, guided hands-on exercises, and group discussions where you synthesize AI-generated insights with your own architectural judgment.

The Trainer

Over the past 25 years, Indu has worked across various industries, including publishing and media, healthcare, finance, biotech, and emergency services. She helped lead and successfully deliver a complex legacy modernization initiative for the NYTimes. She is currently working as a Principal Engineer at CircleCI to help improve the overall developer experience by enhancing CI/CD capabilities and enabling faster, more reliable software delivery.

She loves making sense of complex systems. She leans heavily on Domain-Driven Design and Systems Thinking to unpack complexity in a structured manner, but also borrows freely from Design and other disciplines, to shape unique approaches for understanding and solving problems. She writes about what she learns (and sometimes what she unlearns) at domainanalysis.io.

What People Say About Indu’s Workshops

Indu has a knack for bringing DDD and alike from a buzzword level to truly relatable and actionable knowledge and advice. I have been able to apply what I learnt in Indu’s workshop right away where the rubber meets the road.

Xin Yao, Independent DDD Consultant and Sociotechnical Architect

I was introduced to many new subject matters including service blueprints, liberating structures, and how to lead collaborative workshops to plan and kick off modernization journeys. Highly recommend for any leaders at any stage of a modernization journey!

Jude Bowman, Technical Engineering Lead

FAQ

What level of experience do I need?

This is an intermediate-level workshop. You should be comfortable reading code in at least one mainstream language (Java, C#, TypeScript, Python). The exercises involve reading code, not writing it.

Should I bring my laptop?

Yes. You should already have Claude Code or the AI tool of your choice installed on your laptop. We won't cover installation steps or "what is an LLM."

Is this a workshop about writing code with AI?

No. We don't cover AI-assisted code generation, test writing, or refactoring. The focus is on using AI to understand systems and make better architectural decisions.

Workshop Details

Time: 9:00 am to 5:00 pm each day

Max participants: 25

Prerequisites: Comfort reading code, your laptop with an AI coding tool installed

Included: Coffee breaks and lunch with the trainer and fellow attendees are included in the workshop ticket.