Fraud Detection Data Scientist (AI)

Location: On-site in Almaty, Kazakhstan; relocation support includes a flight and temporary housing.

Company

hirify.global is a generative AI company building AI-powered video creation and next-generation creative tools.

What you will do

• Own the account-abuse scoring system, including signals, weights, thresholds, scoring jobs, monitoring, and recalibration.
• Build early detection for account abuse by combining entity linkage with usage economics and compute costs.
• Detect coordinated abuse rings using payment instruments, IP and ASN concentration, email patterns, registration timing, behavioral fingerprints, and automation signals.
• Design graduated and reversible enforcement actions, including throttling, rate limits, verification, holds, manual review, and bans.
• Measure COGS avoided against lost revenue and support costs, and run legally appropriate holdout tests.
• Work with Payments, Support, and Legal on fraud signals, disputes, appeals, and Terms-of-Use enforcement.

Requirements

• Experience building abuse, fraud, or trust-and-safety detection systems that took action on real users.
• Strong SQL and Python skills for raw event, payment, usage, and identity data; experience with BigQuery is required for the role.
• Understanding of adversarial modelling, censored labels, model decay, and adaptation by abuse operators.
• Experience with graph and linkage methods, entity resolution, connected components, and shared-attribute clustering.
• Precision-first evaluation skills, including cost-weighted errors, operating-point precision, and false-positive budgets.
• Clear written and spoken English at B2+ level.

Nice to have

• Background in fintech, payments, marketplaces, gaming, crypto, trust and safety, platform integrity, or payment operations.
• Experience with gradient boosting, Stripe Radar, chargebacks, disputes, or promotional-abuse modelling.

Culture & Benefits

• High-autonomy role in a fast-moving, flat-structure environment.
• Competitive salary paid in USD.
• Participation in the company stock option program.
• Relocation support with a flight and temporary housing.
• Paid one-month on-site trial.

Hiring process

• 30-minute screening call followed by a 30-minute hiring-manager screen.
• One-hour business case, a practical seven-day home task, and a 60-minute team interview.
• Paid one-month on-site trial.