From Copilots to Orchestrators: A Three-Month Playbook for Training AI-Native Engineering Teams

Most engineering teams are stuck treating AI as autocomplete. Engineers have GitHub Copilot installed (or Claude or Cursor or whatever), they're generating snippets faster, but leaders can't connect usage to business outcomes—and developers are shipping code they don't fully understand. The problem isn't the tools. It's the absence of intentional training.

This talk presents a structured, measurable approach to transform your engineers from "coders with copilots" into AI-native orchestrators who amplify their critical thinking rather than erode it. Drawing from a three-month training pilot launching at a global streaming platform you'll see a bi-weekly cadence of concrete activities designed using proven change management principles that shift AI usage across the full software delivery lifecycle.

What You'll Learn

• Why training isn't optional: The hidden cost of ad-hoc AI adoption—skill erosion, "vibe coding" at scale, and unmeasured experimentation that leadership can't justify
• A change framework that works: How to address vision, skills, incentives, resources, and action plans together using the Lippitt-Knoster model to avoid the confusion and resistance that plague most AI rollouts
• Practical activities you can run Monday: Context engineering exercises, agent-assisted design reviews, pair programming with AI, voice-based prompting experiments, and deliberate "no AI" days that surface where tools help versus where they mask gaps
• Metrics that matter to leadership: How to tag and track AI-assisted work in GitHub, measure PR cycle time and remediation rates, and capture developer confidence through lightweight surveys that reveal ROI

This session is designed for engineering leaders, staff engineers, and managers who need to move beyond "let's see what happens" and build a deliberate training program that turns AI adoption into a competitive advantage. You'll leave with a concrete three-month roadmap—including week-by-week activities, measurement frameworks, and implementation tactics—that you can adapt to your organization starting next sprint.

The core message: 
If you want your entire organization to use AI efficiently, training isn't a nice-to-have. It's the difference between amplifying your team's capabilities and scaling technical debt.