Two enablement teams, working independently, both landed on AI roleplay as the answer to the same problem: knowledge decay after major training moments like SKO. Their solutions look different, and that difference is the point of this session. Rodney's team built Avara ("Coach Kate") for continuous SKO reinforcement, while Dave's team used AI roleplay bots as part of a broader Sales Academy. Run as a joint fireside rather than two separate talks, this is a direct comparison of what each approach solved for, and where they'd each do things differently starting again. Key takeaways: - Two concrete builds solving the same knowledge decay problem, compared side by side - Where each approach trades off cost, complexity, and reinforcement depth - How to decide which model fits your own team's SKO or onboarding structure