ML engineer in trust and safety

AVIV Group runs some of Europe's best-known real estate platforms, and every one of them depends on trust: people need to find a home or a tenant, not a scam or a dead listing. Trust & Safety exists to protect that experience, and our north star is the share of leads that flow from good visitors to good listings. We are hiring a Machine Learning Engineer to build the systems behind it, from identity risk scoring to fraud-ring detection and listing integrity.

• Design, train and ship identity risk scoring, fraud and fraud-ring detection, listing integrity scoring, duplicate and republication detection, and behavioral signals on both sides of the marketplace.
• Build models and retraining loops that hold up as bad actors adapt, catch bad listings before they are reported, and treat false positives as seriously as false negatives, because a false positive is an honest owner who can't publish their home.
• Run careful experiments that account for adversarial drift and feedback loops, where enforcement changes tomorrow's training data, and give an honest analysis when offline and online results disagree.
• Connect reviewer decisions to model training and route uncertain cases to reviewers, integrating with moderation tooling such as Implio and measuring the whole loop.
• Ship integrity and reputation scores that are planned inputs to search ranking and recommendations.
• Own serving, monitoring, SLOs and a share of on-call for what you ship, build the pipelines and evaluation setup the team needs while the group ML/AI platform matures, and strengthen the team through design reviews, mentorship and writing that makes good decisions repeatable.

• Experience shipping at least 3 ML systems to production at meaningful scale, typically with around 6+ years of engineering experience for Senior and 8+ for Staff.
• Hands-on experience in an adversarial or integrity domain such as fraud detection, anti-abuse, spam, content moderation, risk scoring or account security.
• A solid grounding in imbalanced, high-stakes classification: precision/recall trade-offs with real cost on both sides, threshold and policy design, and calibration.
• Experience evaluating models in feedback-loop systems, understanding how enforcement actions distort future training data and having worked around it in practice.
• Strong engineering fundamentals: production-quality code, reliable data pipelines and reproducible work.
• Technical influence beyond your own output, plus clear communication about uncertainty with engineers, policy and operations partners, and leadership, including when the honest answer is "we don't know yet."
• Будет плюсом: graph-based detection (fraud rings, identity graphs, device and payment linkage); LLM-based approaches to trust work (content understanding for moderation, scam-pattern detection, LLM-assisted review); experience alongside human review operations (uncertainty-based routing, active learning from moderation queues); a marketplace or classifieds background; familiarity with EU platform regulation such as the DSA

• Условий в вакансии нет