Senior AI Scientist — Credit Karma (Partner Decision Science)

Intuit; Charlotte (NC); On-site; USD 140,000 - 190,000; Full time

Job summary

Credit Karma is seeking a Senior AI Scientist to join the Partner Decision Science team in Charlotte, NC. You will serve as the technical bridge between Credit Karma’s decision‑science platform and partner data science and credit‑risk teams, turning objectives into production models that improve conversion and revenue.

You will apply machine learning to credit‑risk, targeting, experimentation, and generative AI to deliver relevant financial recommendations for members, collaborating

Qualifications

• MS in Statistics, Mathematics, Computer Science, Economics, Physics, or related quantitative discipline (or BS with equivalent applied experience).
• Approx. 5+ years in credit‑risk analytics, data science, or risk management at a bank, fintech, or credit bureau, with experience in credit underwriting, model development, or customer valuation.
• Deep, hands‑on understanding of credit data, including bureau attributes, risk scores, and alternative data sources.

Responsibilities

• Serve as the technical bridge between Credit Karma’s decision‑science platform and partner banks and fintech lenders’ data science and credit‑risk teams, ensuring partner data and modeling requirements are met end to end.
• Design, build, and optimize large‑scale targeting and approval models that grow conversion and revenue while improving member approval certainty.
• Advance the partner modeling toolkit – including feature engineering on consumer credit, bureau data, modern ML methods, and emerging generative AI applications – to improve partner model performance and member relevance.
• Build strong relationships with partner data science and risk teams, coaching them on the capabilities of the Credit Karma platform.
• Partner cross‑functionally with internal teams to translate partner needs into platform capabilities and bring models into production.
• Represent the Partner Decision Science team in partner‑facing meetings, translating complex technical subject matter for executive and non‑technical audiences and advocating for partners inside Credit Karma and with partners.
• Act as an owner: identify high‑impact opportunities across the partner portfolio, propose solutions, and see them through to production and measurable results.

Skills

Python

SQL

Credit-risk analytics

Model development

Communication skills

Education

MS in Statistics/Math/CS/Economics/Physics

BS with equivalent experience

Tools

BigQuery

ML frameworks

Job description

Overview

Credit Karma is looking for a Senior AI Scientist to join our Partner Decision Science team – the group at the center of how the country's largest banks and fintech lenders build, target, and approve on the Credit Karma Lightbox platform. In this role, you will serve as the technical bridge between Credit Karma’s decision‑science platform and partner data science and credit‑risk teams, turning partner objectives into production models that increase conversion, revenue, and member approval certainty.

This is a highly cross‑functional role. You will partner closely with internal business development, engineering, product, data, and analytics teams, and external partner teams ranging from analysts to senior leaders. Success depends on influencing and translating across audiences as well as technical depth. We want a scientist who can dive deep in data and models, then explain the “so what” to business and partner stakeholders.

You must apply machine learning expertise to credit‑risk and targeting modeling, recommendation and ranking, experimentation, and generative AI to deliver the most relevant and best‑timed financial recommendations for members.

Responsibilities

• Serve as the technical bridge between Credit Karma’s decision‑science platform and partner banks and fintech lenders’ data science and credit‑risk teams, ensuring partner data and modeling requirements are met end to end.
• Design, build, and optimize large‑scale targeting and approval models that grow conversion and revenue while improving member approval certainty.
• Advance the partner modeling toolkit – including feature engineering on consumer credit, bureau data, modern ML methods, and emerging generative AI applications – to improve partner model performance and member relevance.
• Build strong relationships with partner data science and risk teams, coaching them on the capabilities of the Credit Karma platform.
• Partner cross‑functionally with internal teams to translate partner needs into platform capabilities and bring models into production.
• Represent the Partner Decision Science team in partner‑facing meetings, translating complex technical subject matter for executive and non‑technical audiences and advocating for partners inside Credit Karma and with partners.
• Act as an owner: identify high‑impact opportunities across the partner portfolio, propose solutions, and see them through to production and measurable results.

Qualifications

• MS in Statistics, Mathematics, Computer Science, Economics, Physics, or related quantitative discipline (or BS with equivalent applied experience).
• Approx. 5+ years in credit‑risk analytics, data science, or risk management at a bank, fintech, or credit bureau, with experience in credit underwriting, model development, or customer valuation.
• Deep, hands‑on understanding of credit data, including bureau attributes, risk scores, and alternative data sources.
• Expert proficiency in Python and SQL, with experience in modern ML frameworks and comfortable working on large‑scale data (e.g., BigQuery).
• Proven ability to explain complex modeling concepts to non‑technical and executive stakeholders and connect model performance to business outcomes.
• Exceptional verbal and written communication skills, and influence senior audiences in decision‑making.
• External partner‑facing experience is a plus.

Benefits

Competitive compensation package with performance‑based rewards, equity, cash bonus, and benefits. Payment is based on knowledge, experience, and location, with mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandatory mandate mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand mand nom.