Asks for

The ad says“Design and analyze A/B tests, quasi-experiments, and causal inference studies to evaluate the impact of product and policy changes”
The ad says“Design and analyze A/B tests, quasi-experiments, and causal inference studies to evaluate the impact of product and policy changes”
The ad says“Design, build, and maintain scalable data pipelines, monitoring infrastructure, and dashboards to track portfolio health, user behavior, and key risk indicators”
The ad says“Strong understanding of DeFi protocol mechanics (lending, yield vaults, ERC4626) and experience with onchain data tooling (Dune, Shovel, Ponder, Goldsky or similar)”
The ad says“Blockchain data and indexing tools (Dune, Shovel)”
The ad says“Monitor credit risk models, including underwriting, loss forecasting, and fraud detection, and iterate based on observed portfolio performance”
The ad says“Grafana/Prometheus/Metabase”
The ad says“Monitor credit risk models, including underwriting, loss forecasting, and fraud detection, and iterate based on observed portfolio performance”
The ad says“Grafana/Prometheus/Metabase”
The ad says“Strong understanding of DeFi protocol mechanics (lending, yield vaults, ERC4626) and experience with onchain data tooling (Dune, Shovel, Ponder, Goldsky or similar)”
The ad says“Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation”
The ad says“Grafana/Prometheus/Metabase”
The ad says“Python”
The ad says“Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation”
The ad says“Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation”
The ad says“Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation”
The ad says“Blockchain data and indexing tools (Dune, Shovel)”
The ad says“SQL”
The ad says“Monitor credit risk models, including underwriting, loss forecasting, and fraud detection, and iterate based on observed portfolio performance”
The ad says“Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation”

Traditional credit was built for people who already have money. Requirements for credit history, collateral, and costly underwriting create insurmountable barriers for those who need capital most. Over 1.4 billion people lack access to credit. A vendor in Lagos earns cash daily but can't prove a steady income. A Colombian nurse with years of perfect informal repayments remains invisible to banks. Most lending systems spend a lot to guess who will repay, and yet so many who are creditworthy still can’t get a loan.

We built an alternative called Credit. Since December 2024, it has issued over one million unsecured loans using stablecoins. People from around the world have used these loans to pay for things like groceries, medicine, and transportation. Backed by $6.6 million from Paradigm and Nascent, we're scaling a system that has already reached more than 900,000 unique borrowers. Help us take it to the next level.

About the role

We're looking for a data scientist to drive credit risk intelligence across Credit, our leading unsecured lending system. You'll own portfolio monitoring and reporting, research emerging risk trends, and transform borrower behavioral data into actionable guidance that shapes our credit strategy and roadmap.

While our engineering & research teams owns the underlying models, you'll be the person who makes sense of what they're telling us, tracking portfolio health, identifying issues early, and turning insights into clear recommendations for risk strategy and underwriting policy. Over time, this role may expand to drive broader product analytics across our suite of products.

This role is based in San Francisco, California. We work in a hybrid model, with the team in office 3 days per week.

Stack

Python
SQL
Grafana/Prometheus/Metabase
Blockchain data and indexing tools (Dune, Shovel)

Key responsibilities

Monitor credit risk models, including underwriting, loss forecasting, and fraud detection, and iterate based on observed portfolio performance
Design, build, and maintain scalable data pipelines, monitoring infrastructure, and dashboards to track portfolio health, user behavior, and key risk indicators
Partner with product, research, and engineering teams to define north star metrics and translate them into measurable, actionable credit and growth strategies
Design and analyze A/B tests, quasi-experiments, and causal inference studies to evaluate the impact of product and policy changes
Produce portfolio monitoring and investigative analyses, making recommendations based on findings
Translate complex quantitative findings into clear, compelling narratives for product, leadership, and cross-functional stakeholders

Requirements

4+ years of experience in decision science, credit risk analytics, or a closely related quantitative role within fintech or consumer lending
Deep proficiency in Python and SQL; comfortable owning analyses end-to-end from raw data to recommendation
Strong understanding of credit risk modeling concepts, including PD/LGD modeling, scorecard development, reject inference, vintage analysis, and risk segmentation
Demonstrated experience monitoring credit risk metrics and portfolio performance, including loss forecasting and underwriting model improvement
Proven ability to influence and collaborate with cross-functional teams and senior stakeholders, with a track record of translating analytical findings into accessible, actionable insights
Experience designing and evaluating experiments (A/B tests, holdout groups, or causal inference frameworks) in a consumer product context
Comfortable with ambiguity and biased toward action; thrives with minimal oversight and brings strong problem-solving skills and sharp attention to detail

Nice to have

Experience building or maintaining large-scale data pipelines supporting B2C financial products
Familiarity with credit bureau data, cash flow underwriting, or alternative data sources in credit model development
Experience working in emerging markets, ideally on financial products serving everyday consumer needs (microfinance, BNPL, digital lending)
Strong understanding of DeFi protocol mechanics (lending, yield vaults, ERC4626) and experience with onchain data tooling (Dune, Shovel, Ponder, Goldsky or similar)
Exposure to regulatory frameworks relevant to consumer credit (FCRA, ECOA, or equivalent)

Divine Research is an equal opportunity employer.