below marketmost similar roles pay hereabove market
This role pays more than
67%
of similar roles. Most pay
$187,762–$254,750
— the blue band above.
At the midpoint, this role pays about
$246k
versus about
$221k
for comparable roles.
Based on 240 similar postings.
Employer
About Intuit
Intuit is a financial software company known for products like TurboTax, QuickBooks, Mint, and Credit Karma, helping consumers and small businesses manage their finances and taxes. Industry: Financial Software & Technology
Intuit currently has
256 open roles
on FindRole.
Listed pay typically runs
$202,500–$274,000
across 237 roles with salary data.
JOB TITLE: Staff AI Scientist (Fintech Risk AI/ML)
As a Staff AI Scientist in the Money Movement Risk AI/ML team, you will serve as the technical owner for specific money movement risk capabilities. You will design, train, validate, and ship models from first principles to manage real-time decisions regarding card and ACH payments, funding, and settlement. Your daily work involves solving complex statistical problems like extreme class imbalance, delayed labels, and selection bias while building for adaptive adversaries using graph, sequence, and anomaly-detection methods. You will utilize Python, SQL, Spark, gradient-boosted trees, and deep learning to create production-grade features and models. This role addresses the critical business problem of minimizing fraud and credit losses while ensuring small businesses receive their earned funds quickly and reliably.
Design, train, validate, and ship production-grade ML models for real-time money movement risk decisions.
Define and own business outcomes including loss rates, approval rates, and time-to-funds metrics.
Solve complex statistical problems involving delayed labels, extreme class imbalance, and selection bias.
Develop models to detect fraud rings, account takeovers, and evolving attack patterns using graph and anomaly detection.
Build reusable, versioned features from transaction, account, and network data for team-wide use.
Establish offline and online evaluation standards, experimentation frameworks, and adversarial testing protocols.
Partner with ML engineers to implement real-time features, low-latency serving, and automated retraining.
Mentor scientists and engineers while leading design reviews and technical hiring.
What we're looking for
8+ years of experience building and deploying ML models in production, specifically in fraud, payments, credit, or other adversarial financial-risk domains.
MS or PhD in Statistics, Computer Science, Applied Mathematics, Operations Research, or a related field, or equivalent practical experience.
Proven track record of owning models that made real-time, high-stakes decisions with measured business impact.
Deep expertise in gradient-boosted trees, deep learning, anomaly detection, and statistics for calibration, imbalance, delayed labels, and selection bias.
Proficiency in Python and SQL, with hands-on experience in large-scale feature engineering using Spark or similar tools.
Experience shipping models with ML engineers, including real-time features, serving, retraining, and monitoring.
Sound experimental judgment in risk settings, including A/B tests, holdouts, and understanding attacker exploitation.
Clear communication skills to explain trade-offs to risk, product, and business leaders.
Payments risk experience in card and ACH authorization, funding, chargebacks, returns, or settlement (preferred).
Graph ML, network analysis, sequence learning, or experience with LLM and agentic systems (preferred).
Experience with model risk management, explainability in regulated environments, and technical leadership/mentoring (preferred).
Python
R
Scala
Machine Learning
Generative AI
Spark
NoSQL
ETL
Data Warehousing
Statistical Modeling
GLM
Logistic Regression
Decision Trees
Time-series Forecasting
Autoencoders
GANs
Graph Data
Machine Learning
Deep Learning
Python
SQL
TensorFlow
PyTorch
AWS
GCP
MLOps
Apache Airflow
Vertex AI
Scala
Data Pipelines
Natural Language Processing
Time Series
Reinforcement Learning
Graph Modeling
Anomaly Detection