Simon Willison Reports Professional Engineers Are Succumbing to Vibe Coding

Simon Willison

May 7, 2026

Simon Willison observed that the distinction between casual vibe coding and professional agentic engineering is disappearing as AI agents become consistently reliable. He warns that even veteran developers are treating agents as black boxes, shifting the primary software bottleneck from writing code to design and verification.

Simon Willison, creator of Datasette and co-creator of Django, documented a shift where the boundary between Karpathy's vibe coding definition and professional agentic engineering patterns has blurred. As agents like Claude Code prove reliable, even experienced engineers are skipping line-by-line reviews for routine production tasks.

This shift signals a "normalization of deviance" where repeated success without oversight creates a false sense of security. Because AI can now instantly generate professional-looking repositories, traditional engineering signals are losing value. The only remaining proof of software quality is proven usage over time, rather than traditional engineering artifacts.

Treat AI agents like a separate internal team: trust their output as a semi-black box until bugs or performance issues require a deep dive. As coding costs drop, focus on Andrew Ng's strategic design shift. Experience remains the primary amplifier for these tools, making senior oversight critical.

View the full update on simonwillison.net

Simon Willison

@simonwMay 6

I was talking with @josephruscio on the @heavybit podcast the other day when I realized that vibe coding and agentic engineering have started to blur a bit in some of my work - I published some extracts from the transcript https://t.co/evxoG06Vpa

1086

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Keep reading

Andrej Karpathy Outlines the Shift From Vibe Coding to Agentic Engineering

Andrej Karpathy detailed a transition toward an agent-native economy where LLMs serve as the primary substrate for computing rather than just tools for acceleration. He introduced agentic engineering as a high-ceiling discipline for building reliable autonomous systems that replace classical code with natural language instructions.

Simon Willison's Guide Argues Coding Agents Can Raise Code Quality

Simon WillisonMar 18

Simon Willison's Guide Argues Coding Agents Can Raise Code Quality

Shipping worse code with AI agents is a choice. Simon Willison's Agentic Engineering Patterns guide shows how teams can use agents to tackle refactoring debt — API redesigns, concept renames, module splits — that never felt worth the manual effort.

Andrew Ng Outlines Organizational Shifts for AI Native Engineering Teams

Andrew NgApr 28

Andrew Ng Outlines Organizational Shifts for AI Native Engineering Teams

Andrew Ng argues that agentic coding has shifted the primary software development bottleneck from writing code to strategic decision-making. This transition forces engineering teams to adopt lower product manager ratios and requires developers to expand into design and marketing roles.

Vercel Shares Engineering Framework for Shipping Agent-Generated Code Safely

Guillermo RauchMar 31

Vercel Shares Engineering Framework for Shipping Agent-Generated Code Safely

Vercel released its internal guidance for agenting responsibly after shifting to a workflow where AI agents perform the majority of their coding. The framework moves beyond traditional CI testing to include executable guardrails and autonomous deployment rollbacks that contain the risk of AI-generated errors.