Senior Backend Engineer - Search & Discovery (all genders)

Description

The company is ABOUT YOU's Search & Discovery team, which builds the search engine for an e-commerce fashion platform.

The company is one of Europe's fastest growing e-commerce businesses, with offices in Hamburg and Berlin, and works with flat hierarchies and direct communication.

Responsibilities

• Build the search engine in Node.js/TypeScript, covering query understanding, entity resolution and semantic vector search.
• Combine OpenSearch relevance, embedding similarity and behavioral re-ranking signals into one set of search results.
• Create search index pipelines for entities, synonyms and k-NN product vectors that stream data from BigQuery into OpenSearch.
• Integrate LLMs and embedding models such as Gemini and Vertex AI while meeting tight latency budgets, using caching and fallback layers.
• Protect search relevance with golden lists, regression suites and ongoing A/B experiments.
• Help shape the architecture together with the Tech Lead, working closely with data analysts, engineers and product managers.

Requirements

• You enjoy finding out why search results turn out the way they do and turning those findings into concrete technical fixes.
• 5+ years building scalable, low-latency backend APIs. Node.js/TypeScript preferred; engineers who can quickly switch from Java, Go or Kotlin are also considered.
• Deep hands-on knowledge of Elasticsearch or OpenSearch, including mappings, custom analyzers, percolators and scoring functions, beyond basic client use.
• Strong grasp of ranked data: merging, deduplication, score blending and latency trade-offs.
• You want to bring backend engineering and ML together by serving model scores, embeddings or LLM features in production.
• Fluent English and a pragmatic, data-driven approach.
• Working knowledge of IR and NLP basics: tokenization, multilingual search and relevance evaluation frameworks.
• Experience with vector search, such as k-NN/HNSW indices, SigLIP embeddings and tuning recall against latency.
• Experience with LLM engineering, such as prompt caching, timeout budgets and Datadog LLM Observability.
• Familiarity with BigQuery, dbt or Dagster pipelines is a plus.

Conditions

Apply online through the career page. The team responds as soon as possible.

The company works to be welcoming to everyone and values varied backgrounds and perspectives.

Benefits

• Perks are listed on the company's benefits page.