Quelle: CrawlJobs
Für dieses Angebot liegt keine vollständige Übersetzung in der gewählten Sprache vor. Unten sehen Sie den Originalinhalt.

Data Scientist, Antifraud

Higgsfield Almaty, Kasachstan
Full-time On-site work

Publication date: 02/11/2026

Higgsfield AI is a generative AI company focused on AI-powered video creation and next-generation creative tools. This on-site role in Almaty, Kazakhstan is focused on detecting and stopping account abuse, distinguishing fraud from unprofitable pricing on an unlimited generation plan, and building a precision-first, reversible enforcement system that protects paying legitimate customers.

Responsibilities

Own the account-abuse scoring system end to end (signals, weights, thresholds, scoring job, monitoring, recalibration); build early-firing detection that catches abuse before compute is burned; combine linkage (who is connected to whom) and economics (what they consume); treat cost as a gate, not a label; find rings using shared payment instruments, IP and ASN concentration, email-stem families, registration bursts, behavioural fingerprints and automation signatures; work without a reliable device fingerprint and assess its value and cost; design graduated, reversible enforcement actions matched to confidence; own the false-positive budget and appeal path; build pre-grant gates; partner with Payments, Support and Legal; quantify the system's value in money (COGS avoided net of revenue destroyed and support cost created); run holdouts on enforcement where legal and safe; monitor adversarial adaptation and input feature loss; ship scores with reasons; produce evidence usable outside the company; work in SQL, Python and BigQuery on raw event, payment and generation data

Requirements

Experience building abuse, fraud or trust-and-safety detection that took action on real users (payments fraud, promo abuse, multi-accounting, bonus abuse, account takeover, marketplace or gaming abuse); anti-money-laundering or credit-risk backgrounds accepted if the modelling was yours; genuine understanding of the adversarial setting (counterparty responds, training labels come from past decisions, offline AUC on censored labels is close to meaningless); strong SQL and Python on raw event data without help (joins across payments, usage and identity, window functions, cohort replay); gradient boosting and judgment on when a two-condition rule beats a model; comfort with graph and linkage thinking (connected components, shared-attribute clustering); precision-first evaluation instincts (precision at operating point, cost-weighted error); unit-economics literacy (gross margin per account, compute cost); judgment to escalate rather than act when evidence is thin and to say so in writing; clear written and spoken English, B2+

Nice to have

Backgrounds that often do well: fraud/risk data scientists from fintech, payments, marketplaces, gaming or crypto; trust & safety and platform-integrity teams who owned detection and enforcement policy; growth or product data scientists who owned promo-abuse; analysts from payment processors, PSPs or chargeback/dispute operations who moved into modelling

Benefits

Equity participation in the company's stock option program; relocation support (flight + temporary housing); flat structure; high autonomy; fast career growth; competitive salary in USD
Apply with employer

Ähnliche Angebote

Alle →

Cookie-Einstellungen

Jeder Zweck ist optional und standardmäßig deaktiviert. Sie können Ihre Einwilligung jederzeit widerrufen.

Wir zeigen Angebote in Ihrer Nähe