Machine Learning Director

The Role

We’re hiring a Machine Learning Research Manager to lead our ML team and set the technical direction for how machine learning is applied across Arkose’s fraud detection and risk-scoring products. This is a builder-manager role: you’ll grow and coach a small, high-leverage team while staying hands-on with model design, architecture decisions, and code review. You’ll own the connection between fraud outcomes — detection accuracy, false positive rates — and the ML systems that drive them. This role will report to the SVP of Product.

What You’ll Do

• Define and drive Arkose’s Data Science and Machine Learning roadmap across Arkose Titan by aligning research with product priorities.
• Lead development and delivery of ML models and capabilities that power real-time fraud and risk decisioning at Arkose’s scale, in collaboration with detection engineering to turn emerging fraud patterns into features and models.
• Establish modern MLOps practices — model governance, experimentation frameworks, and end-to-end model lifecycle management — and own model performance in production, including precision/recall tradeoffs, drift monitoring, retraining cadence, and latency/cost constraints.
• Drive applied research in graph machine learning, behavioral analytics, and intelligent fraud detection systems — resulting in patents, publications, and product innovation.
• Manage, coach, and grow a team of ML researchers, including performance management, career development, and future hiring.
• Stay hands-on: contribute to model design, feature engineering, architecture reviews, and code/PR review alongside the team.
• Translate fraud and business metrics into measurable ML objectives, and communicate roadmap and tradeoffs to leadership and cross-functional stakeholders.

What We’re Looking For

• 6+ years building and deploying ML models in production, including 2+ years directly managing ML engineers or data scientists.
• A track record of shipping ML systems that moved a real business metric, ideally in fraud, trust & safety, cybersecurity, or another adversarial, imbalanced-data domain.
• Strong technical depth — able to evaluate modeling and architecture choices (classification, anomaly detection, graph-based methods, sequence models) rather than managing from the roadmap alone.
• Experience across the full ML lifecycle: data pipelines, feature engineering, training, evaluation, deployment, and monitoring.
• Proven people-management skills: hiring, coaching, and developing individual contributors.
• Excellent cross-functional communication; able to translate ML trade-offs for non-technical stakeholders.

Nice to Have

• Experience with bot detection, device fingerprinting, behavioral biometrics, or real-time risk scoring.
• Experience with low-latency, high-throughput ML serving infrastructure.
• Familiarity with current ML research relevant to fraud/security (e.g., following or contributing to arXiv publications in adversarial ML, anomaly detection, or graph learning).
• Experience scaling researcher productivity using Agentic AI.