loveholidays, a leading online travel agent operating across eight European markets and processing 60 trillion package combinations daily, is using Codex to change who can build on its technology platform. The company aims to speed up idea-to-product workflows and reduce dependency on a central engineering queue by enabling more people across the business to make code and infrastructure changes.
Who is building and what they build
Product managers, designers and commercial teams are increasingly contributing directly to codebases. They can now prototype customer experiences, make data and infrastructure changes, and turn ideas into working software without every request having to pass through engineering first.
A clear example is Search Playground. Built by loveholidays engineers using the company’s design system, frontend technologies and Codex, Search Playground lets non-engineers create working prototypes, collect feedback and validate whether a new search experience delivers value. More than ten new search experiences have been developed through the Playground; most were built by non-engineers and at least three are already live on the loveholidays website.
One example is Inspire Me, which helps travellers explore different trip types such as beach breaks or foodie escapes. Another came from marketing: for a Crisps from Abroad activation, the marketing team built an interactive microsite in hours using Codex and Search Playground, instead of outsourcing to an agency.
Scaling expertise in data and infrastructure
loveholidays’ Data Platform and infrastructure were originally designed for technical users, so making changes required specialist knowledge and, when people got stuck, intervention from specialist engineers. Codex provides an alternative way to scale that expertise.
Engineering teams encode best practices, validations and instructions into workflows that Codex can guide other users through. Rather than needing to master every underlying system, employees focus on their objective while Codex proposes changes, runs checks and guides the change through the release process.
Measurable results
The company reports substantial improvements over the past year:
- AI-assisted code changes grew from about 7% to 79% in one year.
- Deployments increased by 73% while engineering headcount remained broadly flat.
- Successful AI-assisted changes on the Data Platform rose from 58% to 93%.
- Success across broader self-service infrastructure workflows improved from 63% to 90%.
- The number of Data Platform changes per support request increased fourfold.
There are also direct cost benefits: the Data Engineering team reduced cloud storage costs by roughly £36,000 per year and saved around another £100,000 annually by cutting data-processing waste.
Changing roles and a longer-term ambition
The shift means business teams can act faster without waiting for specialists, and engineers spend less time on routine troubleshooting and more on improving the platform. Dmitri Lerko, Head of Engineering at loveholidays, says: “Everybody is a builder. Making changes to our applications, infrastructure, and deploying code is no longer an engineering-only activity.”
For Mike Jones, CTO, this is part of a broader ambition to build what the company calls the “general intelligence for travel”: combining the company’s technology with employee expertise, and making both more widely accessible through AI.
Lerko describes Codex as becoming a single control plane — a common interface shared by engineers, data scientists and business users. That centralisation reduces the need to teach everyone different tools and makes specialist knowledge available on demand.
Implications for how software is built
loveholidays’ experience points toward a model where specialist expertise is codified and made accessible across an organisation, enabling more people to turn ideas into products and allowing engineers to focus on higher-order problems. As Lerko puts it: “What used to be too hard is now ordinary.” With Codex, the line between the people who have ideas and the people who can build them is beginning to disappear.
