Copilot, Claude, & OptiGPT: The AI Stack Driving Optimizely
From AI-powered PRDs to customer health insights, Cory Liebgott reveals how Optimizely turns everyday workflows into experimentation breakthroughs using AI.
In this episode of LaunchPod AI, Cory Liebgott, VP of Product at Optimizely, shares how her team is using AI in both customer-facing solutions and internal product workflows. From PRD automation to customer insights and emergent AI capabilities, Cory highlights the practical ways product teams can boost productivity without losing sight of trust and governance.
TL;DR
1. Exploring AI tools as a team (00:38)
Cory’s product org regularly hosts “AI show-and-tells” where PMs demo tools like Claude, Copilot, and Cursor. This collaborative culture helps the team quickly separate gimmicks from game-changers.
What you can do: Create recurring team sessions where everyone demos one AI tool. Use this to build a shared toolkit and uncover unexpected use cases.
2. AI productivity in the day-to-day (01:43)
Cory’s team uses Copilot to draft PRDs, refine acceptance criteria, recap meetings, and even translate technical jargon into layman’s terms. For some PMs, Copilot’s voice-to-text feature has been a game-changer.
What you can do: Start small. Use AI to draft specs or summarize meetings. Look for tasks that are repetitive, time-consuming, or require translation between roles.
3. OptiGPT: Optimizely’s internal AI assistant (03:16)
Cory’s team built OptiGPT to surface customer insights, product knowledge, and HR information instantly. It provides exec-style briefs before QBRs and helps onboard new employees faster.
What you can do: Build or adapt an internal AI assistant for customer and product data. Start with FAQs (product docs, policies, usage data) and expand as adoption grows.
4. Emergent capabilities and customer-facing AI (05:10)
Optimizely launched a customer-facing query tool, which unexpectedly became invaluable internally. CS teams now use it to monitor account health and product engagement—sometimes outperforming traditional CS tools.
What you can do: Pay attention to how customers and employees bend tools for new use cases. Double down on emergent wins rather than sticking rigidly to the original spec.
Chapters
00:38 Exploring AI tools as a team
01:43 Real-world AI productivity wins
03:16 Inside OptiGPT: Optimizely’s AI assistant
04:08 Using AI for research and prototyping
05:10 When AI surprises: Emergent capabilities
06:22 Lessons learned: Sharing and experimenting
ICYMI: Why Optimizely is Betting on Agentic AI
Links
What does LogRocket do?
LogRocket’s Galileo AI watches user sessions for you and surfaces the technical and usability issues holding back your web and mobile apps. Understand where your users are struggling by trying it for free at LogRocket.com.