Run · founder · 6 min read

Vibe Coding Is Now on Job Descriptions: What It Means for You

Coinbase, JPMorgan, NVIDIA, and others list vibe coding as a required skill. What that shift means for founders building teams and PMs managing them.

Published May 11, 2026 ·

hiring team building vibe coding career job market

Something quietly shifted in the last few months. Vibe coding started appearing on real job descriptions — not at AI-native startups you’d expect, but at JPMorgan Chase, NVIDIA, TikTok, and Coinbase. The same week Coinbase laid off 700 employees (14% of its workforce) citing AI-driven efficiency gains, its CEO publicly noted that “non-technical teams are now shipping production code.” That’s not a future trend. That’s a live organizational change affecting hiring decisions right now.

If you’re a founder building a team, or a PM managing one, here’s what the shift actually means — and what you should do about it.

What “vibe coding” means on a job description

When a company lists vibe coding as a required or preferred skill, they typically mean one of three things:

The first is AI-assisted prototyping speed. They want someone who can go from concept to working demo in hours using tools like Lovable, Bolt, or Cursor — not someone who writes pristine code from first principles. The prototype doesn’t have to be production-ready; it has to be real enough to get stakeholder buy-in and iterate from.

The second is prompt fluency. The ability to communicate clearly with an AI coding agent — to write good prompts, review AI output critically, catch hallucinations, and steer an agent toward the right solution — is increasingly distinct from “can write Python.” Companies have figured out that these skills don’t automatically co-occur.

The third is reduced dependence on engineering for early-stage work. Firms restructuring around AI efficiency (Coinbase is the loudest current example, but not the only one) want non-technical functions — product, marketing, operations — to ship their own tooling without always queuing behind engineering.

What the Coinbase story actually signals

The Coinbase layoffs made headlines partly because Brian Armstrong framed them around AI productivity rather than the more typical “market conditions” language. His framing was blunt: non-technical employees are shipping production code, and the org is right-sizing based on that new baseline.

He later walked back the “vibe coding to production” framing under pushback — clarifying that all AI-generated code goes through human review. That’s an important detail. No serious company is genuinely bypassing review. But the directional signal is real: the gap between “technical” and “non-technical” roles is narrowing fast, and companies are reorganizing around that fact.

For founders, the takeaway isn’t that you should eliminate your engineering team. It’s that the job descriptions you write for non-engineering roles should probably now include expectations around AI tool fluency — and that some work you’d previously have routed to engineering might reasonably sit with a product-capable founder or PM.

What this means if you’re hiring

If you’re early-stage and hiring a first non-technical employee — an operations lead, a growth person, a content hire — adding “comfortable using AI tools to build internal tooling and prototypes” to the job description is reasonable now. The candidate pool exists. Platforms like VibeCodeCareers and Vibehackers.io are specifically matching these candidates to roles, and analysis of active listings shows the skill appearing in job postings at Fortune 500s, not just startups.

What you should not do is conflate vibe coding skill with the ability to own a production codebase. The most common failure mode is a non-technical hire who builds something impressive in Lovable, ships it as a real product feature without engineering review, and creates security or scalability debt that costs three times as much to clean up. (The site already has a security guide and a data exposure breakdown if you want to understand the risk surface.)

The skill you’re hiring for is more like “AI-fluent operator” than “developer.” These people can prototype, automate internal workflows, and ship low-stakes internal tools. They shouldn’t be your path to production user-facing features without proper review.

What this means if you’re a PM

The Hacker News thread “Ask HN: Is vibe coding a new mandatory job requirement?” surfaced a range of perspectives, but the consensus among working PMs was unsurprising: it’s not mandatory yet everywhere, but it’s increasingly expected at forward-thinking companies, and being unable to demo a working prototype of your own idea is becoming a visible disadvantage.

The practical implication is that PMs who can spin up a working proof of concept — not just a Figma mockup — are increasingly able to get ideas funded, prioritized, and shipped faster. The ability to show a working thing rather than describe it is a different kind of influence than most PM training has equipped people for.

If you’re a PM reading this: start with Lovable or Bolt for web products, or OnSpace AI if you need to demo something on mobile. Get comfortable describing a feature in enough detail that an AI agent can build a plausible version. You don’t need to understand the code it generates — you need to be able to assess whether the output does what you intended.

The emerging job category to watch

The job boards surfacing right now — VibeCodeCareers, Vibehackers.io, vibecoding.work — are tracking a specific emerging profile: the “AI operator” or “vibe engineer.” Not a traditional developer. Not a traditional PM. Someone who sits at the intersection of product thinking and AI tool fluency, capable of shipping their own lightweight tooling and automating workflows that previously required an engineering ticket.

Salaries on Ziprecruiter for “vibe coding jobs” are currently ranging $15–$43/hr, which reflects the wide spread of what’s being asked for — everything from contractor work to full-time roles with real product ownership. The job category is messy and poorly defined right now, which usually means there’s an advantage to hiring early before it gets commoditized and expensive.

The short version

Vibe coding is real enough that it’s showing up in job descriptions at major companies. The core skills being sought are AI tool fluency, prototype speed, and the judgment to know when to escalate to engineering rather than ship AI-generated code directly. As a founder, update your non-technical job descriptions to include these expectations. As a PM, get hands-on with at least one AI builder this month — it’s now a visible career advantage.

The org chart is changing. The hiring expectations are changing faster than most people are adjusting.