Most enterprise developers are stuck in the same place with AI coding agents. They get good results on small, isolated tasks, and nothing dependable on the work that actually matters.

Your coding agent might write a clean method. But it cannot make a change that crosses six projects and a service boundary in the way you expect. This isn’t a limitation of the model. What’s missing is the system around the agents. And building this system is the real work of agentic SDLC and AI software factories.

In this full-day hands-on lab, you’ll lay the foundations for building your own AI software factory. You’ll construct it stage by stage, with each stage unlocking capabilities that are holding you back today. Bring the coding agent your team already uses, whether that's GitHub Copilot, Claude, Codex, or another. The factory is built from practices and patterns, not features of a single tool, so everything you build carries over when you switch agents or use several at once.

Software development in the enterprise is rarely purely brownfield or purely greenfield, and this workshop reflects that reality. You'll add enhancements, bug fixes, and new features on a realistic legacy .NET solution. You'll also stand up a brand new service with new capabilities. By the end of the day, your agentic software factory will take a work item to production, without needing you to do every step manually. Your development process will get streamlined with AI and agentic SDLC.

Along the way, we'll zoom out from the code to the whole lifecycle, from an idea to a feature running and operating in production. When the cost and effort of turning an idea into working software drops this sharply, the bottlenecks move. We'll look at where agentic SDLC fits with, and where it pushes against, the waterfall, agile, and scrum practices your teams already run on. And we'll get specific about which engineering fundamentals matter even more when agents write the code, and which practices and ceremonies have stopped earning their keep.

Agenda highlights

• The agentic SDLC end to end: from idea to production and operations, and where agents fit at each stage
• The new economics of software: what gets cheaper, what doesn't, and where the bottlenecks move
• Sprints, story points, and stage gates when code is cheap: what still works and what to rethink
• Context engineering: giving agents your conventions, frameworks, libraries, and rules
• Specs and decomposition: breaking work down so an agent can verify its own output
• Verification: testing and evaluating results that vary on every run
• Guardrails: bounding what agents can do and what they can trust
• Brownfield: enhancements and bug fixes on a legacy .NET solution
• Greenfield: standing up a brand new service
• Automation: closing the loop from work item to production
• Engineering fundamentals: what matters more, what matters less, and what to let go

Attendee Workstation Requirements:

• You must provide your own laptop for this hands-on lab
• All other requirements will be posted two (2) weeks prior to the conference