Transforming the SDLC with AI
Overview
Transforming the SDLC with AI is a one-day hands-on course that explores how artificial intelligence is reshaping the software development lifecycle from hands-on coding toward human roles focused on system design, agent orchestration, governance, and architectural oversight.
Participants examine how agentic AI tools — including emerging frameworks such as the Agentic AI Foundation Goose project — are changing requirements definition, development workflows, testing, and deployment practices, while also addressing the risks, limitations, and hype surrounding AI-driven development.
Through lectures and practical labs, the course emphasizes human-in-the-loop supervision, responsible automation, and realistic adoption strategies so teams can leverage AI to accelerate delivery without sacrificing quality, security, or accountability.
Duration
2 Days
Who Should Take This Course
Audience
Developers, architects, tech leads, platform teams, engineering managers, other tech leadership
Prerequisites
Attendees will need basic developer skills and familiarity with some traditional software development lifecycle (SDLC) concepts.
Why You Should Take This Course
Upon completing Transforming the SDLC with AI, participants will be able to:
- Explain how AI is reshaping the software development lifecycle, and distinguish between tasks suited for agent automation versus those that require sustained human ownership and judgment
- Design an AI-augmented SDLC that incorporates agentic systems while preserving clear accountability, architectural boundaries, and governance controls
- Apply agentic AI concepts in practice, including supervising and constraining workflows built with Agentic systems
- Orchestrate and oversee AI agents to generate code, tests, and documentation while enforcing human-in-the-loop review and escalation policies
- Identify and mitigate risks introduced by AI-driven development, including security flaws, architectural drift, and hidden technical debt
- Evaluate the evolving role of software engineers, shifting from individual code authorship to system-level design, supervision, and validation of AI-generated systems
Course Outline
Transforming the SDLC with AI
Module 1 — From Human-Centric SDLC to AI-Augmented Systems Design
- Historical SDLC vs. emerging AI-augmented SDLC
- From writing code to specifying intent, constraints and invariants
- Defining architecture boundaries
- Encoding non-functional requirements (security, latency, compliance)
- Reviewing system behavior, not just code
- Where AI works well today and no so well
- Lab: Reframing a Traditional SDLC
Module 2 — The Goose Project
- What “agentic” means
- Core agent capabilities: Planning, Tool use, Memory, Self-reflection and correction
- Why multi-agent systems matter more than “smarter models”
- Overview of the Goose project (Task-oriented agents vs. conversational LLMs)
- Tool-calling and workflow execution
- Lab: Agent workflow for Code generation, Refactoring and Documentation
Module 3 — Humans as Agent Orchestrators
- The agent orchestrator role
- New skills replacing line-by-line coding (Constraint design, system decomposition, policy-driven development)
- Oversight models (Review-everything, Trust-but-verify, Tiered autonomy)
- Architectural patterns for AI-generated systems (Bounded contexts, Contract-first APIs, “AI-unsafe” zones)
- Lab: Designing an Agent-Orchestrated Build Pipeline
Module 4 — Future SDLC: Governance, Risk, and Reality
- Growing importance of governance
- AI-generated code and security vulnerabilities, license contamination and hidden technical debt
- Auditing agent decisions vs. auditing humans
- Measuring success with time-to-validated-system
- New “AI systems engineer” roles
- AI Concentrates responsibility
- Lab: Stress-Testing an AI-Transformed SDLC