AI Transformation at Scale: Practical Adoption Across 150+ Engineers with Peter Gostev // Head of AI @ MoonpigNavigating AI hype versus implementation reality in large engineering organizations
Peter Gostev · Head of AI
Plays in the Podigee player. Podigee may set cookies after you press play.
How do you drive meaningful AI transformation across 150 software engineers without mandates or force? Peter Gostev, Head of AI at Moonpig, reveals the technical strategies and organizational approaches behind scaling AI adoption from 130 to 400+ users while navigating the gap between industry hype and implementation reality.
From managing complex integration challenges where building AI-powered features requires 80-90% traditional software engineering (integrations, scaffolding, deployment, UI) to implementing three-pillar strategies (tool adoption, automation workflows, experimental features), Peter shares hard-earned insights on building AI capabilities through process re-engineering rather than simple automation.
Technical insights for CTOs and engineering leaders:
- 🏗️ Portfolio approach: balancing quick wins with experimental high-impact projects
- ⚡ Prototype-first methodology for validating AI solutions before full development
- 🤖 Reality gap between agentic AI hype and production deployment complexity
- 👥 Organic adoption strategies that scale without top-down mandates
- 🔧 Custom GPT frameworks for non-technical subject matter experts
- 📊 Why most AI work is integration, scaffolding, and deployment—not just AI
- 🔄 Process re-engineering with AI: changing workflows rather than automating existing inefficiencies
Chapters.
- 00:00:51Introduction: Peter Gostev, Head of AI at Moonpig
- 00:03:49First Computer Memory: Pentium 2 and the Golden Age of Learning
- 00:05:28Career Evolution: From Gaming to Data Science and Machine Learning
- 00:07:01AI Hype Cycle: Underestimating Future Potential While Overhyping Present
- 00:09:41Agent Reality Check: Why Full Agentic Systems Are Harder Than Expected
- 00:13:51The Decade of Agents: Learning from Autonomous Driving Timelines
- 00:15:09Developer Tool Adoption: The Skill Gap in AI-Assisted Programming
- 00:18:40Organizational Change: Teaching 150 Engineers New Workflows
- 00:20:51Claude Code Success: Finding Product-Market Fit in CLI Tools
- 00:24:33Context Management: Why Proper AI Usage Requires Intuition
- 00:26:07Three-Pillar AI Strategy: Tools, Automation, and Experimentation
- 00:28:01Team Structure: AI Engineers vs Data Scientists
- 00:30:10Partnership Models: Building vs Enabling Other Teams
- 00:32:51Adoption Management: Direct Collaboration vs Ambassador Networks
- 00:35:45Prototype Power: One Evening of LLM Work vs Weeks of Traditional Development
- 00:38:02Subject Matter Expertise: Custom GPTs for Non-Technical Users
- 00:42:45Writing with AI: Research vs Synthesis and the Importance of Human Thinking
- 00:47:08Tool Strategy: ChatGPT Rollout Without Internal Marketing
- 00:49:58Adoption Reality: 400 Users But Still Only 5-10% of Potential Usage
- 01:02:06Training Challenges: From Individual to National Level Education
- 01:05:10Future of Development: What Changes and What Remains Human
- 01:08:24Developer Productivity: Higher Ambition, Not Less Work
- 01:16:50University Concerns: Are Students Prepared for AI-First Development?
- 01:18:47Career Advice: Do Hard Technical Things Early While You Can
Next up. More from the top of the top.
All episodes →
