Research Scientist II

TL;DR

Research Scientist II (Fraud/ML): Building fraud risk models and a scam detection stack for voice and IVR interactions with an accent on applied machine learning, multimodal signal modeling, and translating research into production-ready detection improvements. Focus on designing reproducible evaluation workflows, analyzing evolving fraud patterns, and recommending practical mitigation steps based on investigation findings.

Location: US Remote

Salary: $160,000 - $185,000 USD

Company

hirify.global provides identity trust and deepfake/fraud detection for voice, video, and digital interactions.

What you will do

• Build and improve fraud risk models and scoring systems using audio, behavioral, and metadata-based signals.
• Analyze fraud patterns across customer environments and translate findings into measurable model and workflow improvements.
• Research and build a scam detection stack from conception to realization.
• Partner with engineering to move successful research into production and improve live fraud outcomes.
• Support high-priority fraud investigations by analyzing system behavior, attack patterns, and detection gaps.
• Design and maintain reproducible research workflows, internal tools, and evaluation pipelines.

Requirements

• Location: Must be based in the US
• Advanced degree (Master’s or PhD) in a quantitative field (CS, math, statistics, engineering, AI) or equivalent applied research experience.
• 3+ years of professional experience in machine learning and fraud detection/risk modeling/speech or signal processing/anomaly detection (or closely related).
• Strong Python skills for research tooling, experimentation frameworks, and model evaluation workflows.
• Hands-on experience with modern ML frameworks such as PyTorch, TensorFlow, or Keras.
• Proven ability to translate research into practical improvements for models, decision systems, or production-facing recommendations.

Nice to have

• Experience with fraud/scam detection in voice, IVR, contact center, authentication, or trust & safety.
• Experience building and/or fine-tuning multimodal foundation models.
• Experience improving precision/recall in real-world detection systems (thresholding, scoring, watchlists, entity-resolution signals).
• Familiarity with metadata-driven risk signals (telephony, carrier, device, account, behavioral indicators).
• Working knowledge of C/C++ or Go.

Culture & Benefits

• Remote-first environment with regular in-person team on-sites and company events.
• Unlimited Paid Time Off (PTO).
• Competitive compensation including RSUs for all employees.
• Health and welfare plans including an employer-paid employee-only option and an HSA employer contribution.
• Paid parental leave and additional family support benefits.
• Annual learning & development stipend and monthly phone/internet allowance.

Hiring process

• Applications may be prioritized using AI tools for human review based on job-fit signals.
• Selected candidates move forward for further evaluation and interviews.