At Monzo, the gap between design intent and shipped code is closing. Traditionally, designers identified UI friction or copy inconsistencies but had to wait for engineering capacity to fix them.
In 2026, Monzo has intentionally shifted this dynamic, empowering designers to ship small, well-scoped changes directly into production code using AI-native tools, while maintaining strict banking safety standards.
The catalyst: “Designer shipping to prod”
The movement was spearheaded by Senior Lead Designer Heldiney Pereira, who championed AI internally and launched a “Designer Shipping to Prod” competition.
This initiative created a framework for designers to move beyond Figma and into real engineering guardrails, ensuring that design quality is treated as a functional requirement for customer trust.
How designers are using AI and engineering guardrails
The process is not “vibe coding” in isolation. It is a structured integration into Monzo’s existing CI/CD pipeline. Designers work in real repositories, commit via GitHub, and go through mandatory code reviews.
Susan Walsh (Lead Product Designer):
Using Cursor, Susan found that fixing small, well-scoped issues directly in production creates more leverage than prototyping tools. For example, she corrected cross-platform copy inconsistencies (e.g., “Get Paid online” vs. “Get Paid”) that would otherwise remain in the backlog.
Sasha Ward (Lead Product Designer):
On the Joint Accounts team, Sasha uses AI to query codebases directly. Even without experience in Go, she has contributed modest backend changes. AI acts as a translator, helping designers understand most of the architecture while relying on engineers for final validation.
Justin Thrift (Senior Product Designer):
Working on a new product in stealth mode, Justin uses feature flags to deploy design updates. This allows him to preserve product details like micro-interactions and refined copy without adding pressure to engineering teams during critical launches.
Key takeaways for design-engineering collaboration
- Tooling: AI-native editors such as Cursor and GitHub
- Safety: Changes are controlled through feature flags and mandatory reviews
- Ownership: Designers take responsibility for outcomes, not just intent
- Productivity: AI reduces onboarding time by explaining complex systems quickly
The goal: Design as responsibility
This shift is not about replacing engineers, but about expanding the role of designers. By operating within established guardrails, Monzo’s design team has moved from experimentation to consistent delivery.
The result is a workflow where product quality and speed are no longer in conflict, and the details that customers notice are not sacrificed for velocity.
