Monnai is building the world’s leading identity and risk data infrastructure, powering some of the largest global digital lenders, financial institutions, and fintech players across the U.S., LATAM, Europe, and Asia. At our core, we focus on data quality, operational performance at scale, and delivering measurable outcomes for customers.
Monnai recently raised a new $10M+ financing round and is operating from a position of strength, having reached profitability and grown revenue 5× in 2025. To date, the company has been deliberately focused on building deep technology along with basic ML Score and strong, hands-on customer engagement and high-leverage product execution.
Operating Environment & What to Expect
Monnai is a real, working business with a proven product, strong customer adoption, profitability, and rapid growth - but we are still early. Teams are lean and hands-on, processes are evolving, and priorities can shift quickly as we learn and scale. Direction is not always fully formed; clarity often needs to be created.
This role is ideal for someone who enjoys working with ambiguity, limited structure, and constant motion - and who can turn rough inputs, evolving ideas, and incomplete direction into something coherent, polished, and distinctive.
At Monnai, we operate as one team, where clarity, customer impact, creativity, and long-term value creation come first. Success isn’t measured by Features or Models count - it’s measured by whether customers and prospects get the real value, are compelled to test the platform, and expand usage over time.
What This Role Is About
We are looking for a Senior Data Scientist to lead the development of next-generation credit underwriting and fraud/identity risk models, leveraging both traditional signals and alternate data (e.g., digital footprint, device/behavioral, transactional, and ecosystem data). You will own an end-to-end model lifecycle—from problem framing and feature innovation to real-time deployment, monitoring, and governance. This role is central to improving approval and conversion while protecting the platform from fraud losses and credit defaults.
You will work closely with Risk, Product, Data Engineering, ML Engineering, Compliance, and Operations to deliver models that are high-performing, explainable, and audit-ready across markets and customer segments.
What You’ll Do
Credit Risk Modeling
- Build and optimize models for credit underwriting, limit assignment, and pricing/risk-based segmentation.
- Develop scorecards and ML models for probability of default (PD), early delinquency, and roll-rate prediction.
- Run cohort/vintage/portfolio analytics to detect macro shifts and segment drift.
Fraud / Identity Risk Modeling
- Develop real-time and near real-time models for:
- application fraud, synthetic identity, account takeover
- transactional fraud and anomaly detection
- Combine ML with risk rules and policy constraints to drive production decisioning.
Alternate Data & Digital Footprint Signals
- Invent, validate, and productionize features from:
- device intelligence, IP/proxy/VPN risk, location consistency
- user behavioral patterns, session velocity, clickstream heuristics
- onboarding/KYC friction signals and ecosystem-linked attributes
- Handle sparse, noisy, or adversarial signals and design strategies for cold-start users.
Experimentation & Impact Measurement
- Design offline evaluation + A/B testing to measure impact on:
- approval rate, default rate, fraud loss, false positives/negatives, unit economics
- Establish guardrails for stability and fairness under changing population dynamics.
Production & MLOps
- Partner with engineering to deploy models into batch + low-latency decision APIs.
- Implement monitoring for drift, stability, calibration, and feature health.
- Drive challenger models, retraining cadence, and automated rollback strategies.
Model Risk Management & Governance
- Build models aligned with regulatory / internal standards:
- explainability (e.g., SHAP-based narratives)
- bias/fairness assessment
- clear documentation of assumptions, limitations, and validation
- Support internal reviews, audits, and compliance reporting.
Leadership
- Lead complex, cross-functional projects with high autonomy.
- Mentor junior DS members; set best practices for rigor and reproducibility.
- Influence risk roadmap and long-term modeling strategy.
Who You Are
- 5–8+ years of data science / ML experience with strong focus on credit risk and/or fraud risk in fintech or lending.
- Expert in statistical and ML methods: logistic regression, GBDTs (XGBoost/LightGBM), anomaly detection, time-series, survival/transition models.
- Strong Python (pandas, sklearn, XGBoost/LightGBM, PyTorch/TensorFlow) and SQL.
- Experience deploying models into production and collaborating with MLE/MLOps teams.
- Proven ability to balance performance + interpretability + compliance.
- Excellent stakeholder communication—able to explain risk outcomes simply and credibly.
Preferred Qualifications
- Experience with alternate data/digital identity, device graphs, behavioral biometrics, or ecosystem risk signals.
- Experience with streaming/real-time systems (Kafka/Flink/Spark Streaming).
- Understanding of fraud adversarial dynamics and robust modeling under attack.
- Cloud + orchestration stack exposure (AWS/GCP/Azure; Airflow/Prefect/Dagster; dbt).
What Success Looks Like
- Credit models increase good approvals while reducing defaults and improving portfolio stability.
- Fraud/identity models reduce losses without harming legitimate users (lower false positives).
- Models are explainable, monitored, and resilient to provider and behavior shifts.
- You independently deliver impactful systems and raise the technical bar for the team.
