Data Scientist (AI Search & Ranking)

Location: Düsseldorf, Germany; hybrid with up to 2 work-from-home days weekly. Remote work from Germany or selected countries abroad is available for up to 20 days per year.

Company

hirify.global operates a hotel metasearch engine that compares prices from hundreds of booking sites using real-time auction, machine learning, and petabytes of travel data.

What you will do

• Own search and ranking components end-to-end, from problem framing and model development through production deployment and measurable business impact.
• Build multilingual query and intent understanding for intent classification, named entity recognition, slot filling, and semantic interpretation across 35+ languages.
• Design candidate generation, dense and hybrid retrieval systems while balancing recall, precision, and query-time latency.
• Develop ranking and personalisation models using in-session and long-term user behaviour, training-data construction, debiasing, A/B testing, and conversion analysis.
• Apply and fine-tune language models for query rewriting, result explanation, and complex queries while managing latency, cost, and quality trade-offs.
• Design offline and online evaluation frameworks, including relevance judgement pipelines, cold-start evaluation, and retrieval and ranking metrics.

Requirements

• 5+ years of experience building and shipping search, ranking, or recommendation systems in production. A Master’s degree or PhD is preferred, or equivalent demonstrated expertise.
• Deep theoretical and practical experience in at least one of learning-to-rank, retrieval and candidate generation, query understanding and NLP, two-tower architectures, or personalisation, plus working knowledge of the others.
• Experience fine-tuning and distilling transformer models for production-scale, latency-efficient solutions.
• Strong experimentation skills, including A/B test design, bias awareness, guardrail metrics, and connecting offline quality with online business outcomes.
• Strong Python and SQL skills; hands-on experience with PyTorch or HuggingFace.
• Familiarity with vector search infrastructure, cloud ML pipelines such as Vertex AI and Airflow, and GCP.
• Clear communication, collaborative mentoring, entrepreneurial drive, ownership, analytical rigour, and comfort working in ambiguous problem spaces.
• Outcome-driven approach focused on business impact, quality, continuous learning, feedback, and practical use of AI tools.

Nice to have

• Experience with travel, e-commerce, or two-sided marketplace search.
• Familiarity with marketplace dynamics and advertiser signals in relevance ranking.
• Multilingual retrieval experience.
• Published research in information retrieval, ranking, or NLP at SIGIR, RecSys, WWW, or KDD.

Culture & Benefits

• Work with petabytes of live travel data, modern ML infrastructure, and meaningful end-to-end ownership in lean teams.
• Models and analyses directly influence traveler experience and marketplace outcomes, with performance measured by impact rather than methodological sophistication.
• Personalized coaching, workshops, educational meetups, conferences, online learning courses, and access to a campus library.
• Flexible working hours, self-determined vacation with a minimum of 25 days, and remote work from Germany or selected countries abroad for up to 20 days per year.
• Visa support, relocation package, interest-free newcomer loan, free language classes, and team and company-wide events.
• Canteen budget, snacks and drinks, on-site gym and sports options, ergonomic workspaces, childcare support, and an on-campus kids room.

Hiring process

• 10-minute one-way interview followed by an introduction call with the hiring manager.
• Technical interview lasting 90 minutes with system design on a real problem, followed by a 60-minute collaboration interview.
• Potential 30-minute final interview with the team lead and a Profile Dynamics Assessment.