Data Analyst, Internal Audit

About the role

GiveDirectly is hiring a Data Analyst for its Internal Audit function in Nairobi. The role uses data to find and reduce fraud risk across GiveDirectly's programs and geographies, working closely with data, product and engineering teams.

Key responsibilities

• Analyze fraud risk trends by mining enrollment, payment, survey and field-operations data to spot patterns, anomalies and emerging fraud typologies such as duplicate enrollments, collusion, identity fraud and diversion of funds.
• Design and oversee fraud-risk indicators, building and validating recurring monitoring with central data and operational control owners.
• Identify data quality issues in data collected by Internal Audit and help design process improvements to fix them.
• Embed data-driven thinking into Internal Audit process design and package findings into briefs and presentations for country teams, operations leadership and senior management.
• Support fraud investigations by providing data pulls, quantifying exposure and helping reconstruct events, with documented queries, data lineage and QA checks.
• Partner with Product and Central Data on fraud-detection tooling, translating risk signals into product requirements and acceptance criteria.
• Partner with Data Engineering on upstream data quality and pipeline improvements, while owning the downstream analytical layer.
• Back-test risk indicators against investigated cases and monitor precision, coverage and false-positive rates.
• Communicate uncertainty honestly, distinguishing confirmed fraud from unconfirmed patterns.

Required profile

• 4+ years of experience in data analysis, audit, risk or a related analytical role.
• Experience in internal audit, fraud/forensics or a nonprofit/development context is a plus but not required.
• Ability to turn exploratory analysis into a clear, defensible narrative for non-technical stakeholders.
• Interest in working cross-functionally with product and engineering teams.
• Sound judgment on evidence quality and high integrity and discretion, given access to sensitive information about staff, recipients and investigations.
• Alignment with GiveDirectly's values, including prioritizing recipient wellbeing.

Required skills

• Strong SQL and comfort with large, messy operational datasets (Salesforce, payments, survey data).
• Python strongly preferred for reproducible analysis, automation, record linkage, anomaly investigation and data-quality testing.
• Visualization tools such as Looker, Tableau or Power BI, and/or Python/R for analysis.
• Familiarity with analytical notebooks, version control and reproducible documentation is a plus.