20Product: Is the Design Phase Dead in a World of AI | Has Claude Code Crushed Anthropic Already | What Roles of a PM Are Less and More Important with AI | How the Best Product Leaders Tell Stories with Noam Lovinsky, CPO @ Superhuman

What It Covers

Noam Lovinsky, CPO at Superhuman, discusses how AI transforms product development workflows, team structures, and the role of product managers. He covers vibe coding, Claude Code adoption, spec writing for agents, and predictions for twenty four seven inference.

Key Questions Answered

• •AI Code Generation Impact: Superhuman approaches 50% of net new code written by AI, targeting 90% within twenty four months. Teams don't shrink but build more products with smaller ratios—shifting from five to ten engineers per PM to just two engineers per product team.
• •Spec Writing Evolution: Product teams now write specifications for AI agents rather than humans, requiring different structure with explicit context libraries, examples of past successes and failures, and using the coding agent itself to help write specs it will later execute for better compatibility.
• •Design Phase Transformation: Design thinking remains essential but shifts form. High fidelity prototypes via vibe coding become standard for stakeholder communication, while early exploration still benefits from whiteboards and Figma. Claude Code dominates as the primary tool, with engineers transitioning from Cursor to terminal based workflows.
• •Product Team Acceleration: AI most significantly shrinks the exploration phase and iteration cycles. Teams test multiple solution variations in parallel, dramatically increasing learning velocity. Engineers spend saved time on product strategy, data analysis, and wearing multiple hats rather than purely reducing headcount or costs.