A journalist recounts shipping code as a vibe coding engineer during a two-day stint at Notion, working alongside engineers using AI tools like Cursor
I spent two days at Notion and saw an industry in upheaval. I also shipped some actual code. — I asked my editors if I could go work at a tech startup.
Context & Ripple Effects
This is an on-the-ground example of AI coding tools moving from product demos into day-to-day engineering workflows: a visitor at Notion was able to participate in shipping code while working beside Cursor users. Later coverage of developers describes a related shift toward feeling more like architects than construction workers.
The account also puts Cursor’s adoption in the context of a company still defining its operating model; a later report on the vendor highlighted Cursor’s intensive internal product-testing culture. The significance here is less a claim about output than evidence that AI-assisted coding is becoming observable workplace practice.
First-order effects
- Notion engineers’ use of Cursor becomes part of the working environment encountered by a short-term, nontraditional contributor, lowering the practical barrier to participating in a code-shipping workflow.
- Cursor gains a concrete enterprise-workflow reference point, while the journalist’s experiment gives readers a firsthand view of how AI tooling changes the entry experience for coding work.
Second-order effects
- Coding-tool rivals face pressure to make their agents usable inside real team workflows, not merely effective in isolated code-generation tasks.
- Engineering managers may put greater weight on review, architecture, and integration practices as more contributors can generate proposed changes with AI assistance.
Third-order effects
- If this pattern persists, software teams may separate code production from accountability more sharply: more people can propose implementations, while experienced engineers retain responsibility for system design and release decisions.
- The broader labor shift is toward AI-mediated knowledge work, where tool access expands participation but raises the value of oversight and organizational context.
The trend: AI coding is evolving from an individual developer aid into an agentic work surface that changes who can contribute to software and what engineers are chiefly accountable for.