Senior Search Backend Engineer (mfx)

Real innovation starts with a strong foundation and at commercetools that comes from the perfect balance of our product and our people.

Behind every leap forward is a collective of builders explorers doers makers and kind of people who not only pioneered a more flexible approach to commerce architecture but also shaped the culture of experimentation that approach unlocked. Together they are the engine of commerce innovation today.

At commercetools we power the next era of commerce for our customers. Whether its AI-driven solutions that help enterprises make smarter business decisions bridging digital and physical shopping experiences or enabling entirely new ways for industries to connect with their customers we help the worlds most ambitious companies experiment scale and grow without limits.

Here the best idea wins not the loudest voice. You will have the tools trust and space to not only build the future of commerce but to build your own.

As a Senior Search Engineer on our Discovery Team you will build the retrieval layer behind product discovery for some of the worlds most demanding commerce brands.

Search is changing shape. Shoppers type full sentences instead of keywords and increasingly the client calling our API is an AI agent rather than a person. We are on the path to take our semantic and hybrid retrieval to general availability and tackling the hard problems: making it fast affordable measurably better across enterprise catalogs and working out what comes after it. You would join that work at every layer from query understanding down to the vector index and the cost of a single query.

You will:

• Design and implement search functionalities in a PaaS ecommerce Scala backend
• Develop search capabilities that combine lexical semantic and hybrid retrieval
• Develop search-related libraries based on ElasticSearch and MongoDB
• Contribute to search relevance and quality and to how we measure it
• Run and maintain a scalable search infrastructure as code on top of Kubernetes using terraform and multi-tenant architecture
• Test software components regarding usability functionality and performance and work closely with Product Management and DevOps
• Learn Scala and functional programming as your new main development tools
• Take part in on-call rotation with a worldwide distributed team for production systems

This role is hybrid with three days a week spent in our Munich Berlin London or Valencia office.

Youre a creative problem-solver who is wired to find solutions. You confidently dive into complex challenges and have a talent for making them simple for others. Your curiosity drives you to constantly grow and contribute to an environment of trust and teamwork.

Great ideas come from many paths and your unique perspective matters more than checking every box. What matters most is the mindset you bring to the work. You do not need to be an ML researcher for this role. You need to be a strong backend engineer who is genuinely interested in retrieval quality.

You bring:

• At least 5 years of experience as a Software Engineer with production experience on the JVM
• Solid understanding of parallel and asynchronous programming as well as non-blocking I/O
• Experience developing REST APIs and knowledge of scalable architectures including sharding replication load balancing and failover
• Interest in working with Scala using pragmatic functional programming patterns. If you come from another JVM language we will support you in picking it up.
• Fluent English for work in international teams
• Experience operating multi-tenant SaaS where one tenants load must never affect anothers
• Knowledge of scalable event-driven architectures including queues and notification topics
• A genuine curiosity for using AI tools to work smarter and more effectively paired with a drive to learn and put them into practice in your role.

Nice to Have:

• Hands-on search experience preferably ElasticSearch or OpenSearch including a working understanding of how an inverted index scores and ranks documents
• Practical exposure to vector retrieval such as embeddings approximate nearest neighbor indexes and hybrid ranking and a sense of where each approach breaks down. A deep lexical search background plus a clear appetite to go into this territory also works.
• Comfort reasoning about relevance with data including precision and recall tradeoffs ranking metrics and behavioral signals
• Experience with infrastructure work and other technologies like Helm Kubernetes and Rust