[Remote] Model Risk Data Scientist at SentiLink

United States

Apply Now

Not SpecifiedCompensation

Mid-level (3 to 4 years)Experience Level

Full TimeJob Type

UnknownVisa

Fintech, Financial Services, Identity VerificationIndustries

Requirements

Bachelor’s, Master’s, or PhD in Data Science, Statistics, Computer Science, or a related field
3 years of work experience in a related technical field, or 5-7 years relevant applied academic experience
Proven experience in data analysis, modeling, and performance evaluation
Strong proficiency in Python, and specifically data analysis libraries (Pandas, Numpy), Data Visualization (Python matplotlib Plots, Excel Plots / BI tools), and SQL
Ability to interpret and communicate complex data insights to both technical and business audiences
Exceptional problem-solving and analytical skills with a focus on actionable results
Interest in developing deep domain expertise for model risk analysis and model governance work
Ability to thrive in a fast paced environment characterized by the need to solve extremely varied, high impact, open ended problems
Proven experience in assessing the quality, stability, performance and behavior of production grade ML models, ideally from the perspective of model governance, fair lending or economic risk (highly preferred)
Familiarity with fraud detection (preferred)
Experience with GitHub
Candidates must be legally authorized to work in the United States and must live in the United States

Responsibilities

Build out foundational processes for generating and surfacing model performance metrics, including crafting + calculating the metrics and building python rails
Act as point of contact for Strategic Financial Partners and manage their model governance needs
Regularly conduct performance and exploratory analyses to establish the quality of model outcomes
Provide guidance on the appropriate use of products, respond to inquiries around model development procedures and generate stability analyses
Create automation to detect inconsistencies or issues with the production models that power fraud detection products, such as data driven gap analysis to automation via anomaly detection or training challenger models to identify weaknesses
Primarily work with a Python ecosystem, using SQL, SageMaker, S3, Metabase and Git to support

Skills

Key technologies and capabilities for this role

PythonSQLSageMakerS3MetabaseGitmodel performance metricsanomaly detectiondata analysis

Questions & Answers

Common questions about this position

Is this position remote?

Yes, this is a remote, US-based role.

What are the required skills for this role?

The role requires a Bachelor’s, Master’s, or PhD in Data Science, Statistics, Computer Science, or related field; 3 years of work experience in a related technical field or 5-7 years relevant applied academic experience; proven experience in data analysis, modeling, and performance evaluation; strong proficiency in Python (Pandas, Numpy), data visualization tools, and SQL; and experience assessing production ML models.

What is the salary or compensation for this position?

This information is not specified in the job description.

What is the company culture like at SentiLink?

SentiLink has a fast-paced environment characterized by solving varied, high-impact, open-ended problems, with rapid growth and a focus on innovative identity and risk solutions.

What makes a strong candidate for this Model Risk Data Scientist role?

Strong candidates have experience assessing production ML models from a model governance perspective, exceptional problem-solving skills, ability to communicate complex insights to technical and business audiences, and interest in deep domain expertise in model risk analysis.

SentiLink

Machine learning solutions for identity fraud detection

About SentiLink

SentiLink provides solutions to help financial institutions prevent identity fraud. Their main product uses machine learning models to detect fraudulent activities during the application process. By analyzing data and reviewing cases with a team of risk analysts, SentiLink offers insights that help clients make informed decisions about approving customers. What sets SentiLink apart from competitors is their ability to adapt their products to various types of fraud and customize them to meet the specific needs of each client. The goal of SentiLink is to enable financial institutions to minimize fraud losses while maintaining a positive customer experience.

San Francisco, CaliforniaHeadquarters

2017Year Founded

$81.7MTotal Funding

SERIES_BCompany Stage

Fintech, AI & Machine Learning, Financial ServicesIndustries

51-200Employees

Benefits

Health Insurance

401(k) Retirement Plan

401(k) Company Match

Unlimited Paid Time Off

Home Office Stipend

Risks

Sophisticated fraud techniques may outpace SentiLink's current detection capabilities.

Dependence on high-quality data could impact fraud detection effectiveness if data quality declines.

Intense competition in the fintech sector may threaten SentiLink's market position.

Differentiation

SentiLink specializes in detecting synthetic identities using proprietary machine learning models.

The company offers customizable solutions that integrate seamlessly with clients' existing systems.

SentiLink's Facets tool provides feature-specific intelligence to enhance fraud detection accuracy.

Upsides

Growing demand for synthetic identity fraud detection boosts SentiLink's market potential.

Partnerships with Scienaptic AI and Persona enhance SentiLink's fraud detection capabilities.

API-based solutions allow SentiLink to offer seamless integration and customization.

Land your dream remote job 3x faster with AI

Try Jobo Free