How loveholidays is making everyone a builder with Codex
loveholidays is using Codex to let non-engineers contribute directly to its codebase, accelerating prototyping and reducing engineering bottlenecks across its travel platform.
loveholidays, an online travel agent operating in eight European markets, processes 60 trillion package combinations daily to help customers book holidays. The company has built a proprietary technology platform to streamline holiday searches and bookings. Now, it is using Codex to enable product managers, designers, and commercial teams to modify code directly, bypassing traditional engineering queues. According to Dmitri Lerko, Head of Engineering, this shift means "everybody is a builder," with stakeholders driving deployments and value creation across the platform.
Previously, non-engineers had to rely on engineering teams to prototype new customer experiences, creating opportunity costs. loveholidays addressed this by developing Search Playground, a tool built with Codex that allows employees to turn ideas into working prototypes without engineering support. More than ten new search experiences have been created, including Inspire Me, a feature helping travelers explore trip options. Marketing teams also used the tool to build an interactive microsite for a competition in hours, avoiding external agency costs and delays.
Behind the scenes, loveholidays’ Data Platform and infrastructure were originally restricted to technical users, requiring specialist knowledge to make changes. Codex now encodes engineering best practices into guided workflows, allowing non-technical employees to propose, validate, and deploy changes independently. Lerko notes that this approach makes the expertise of data and infrastructure engineers "available on tap, 24/7" through Codex, reducing dependency on specialist support.
The results show measurable improvements: AI-assisted changes to the Data Platform rose from 58% to 93% over a year, with four times as many changes per support request. Engineering organization-wide AI-assisted code changes increased from 7% to 79% in the same period. This shift reduces routine troubleshooting for engineers, allowing them to focus on platform improvements while accelerating innovation across the business.