WBG Pioneer -Financial Data Engineering Intern

World Bank Group•Washington, DC

1d•Onsite

About The Position

WBG Pioneers, the World Bank Group’s Internship Program, offers undergraduate and postgraduate students a high impact learning experience at the heart of global development. Participants gain hands on experience in a diverse and dynamic environment, contribute fresh perspectives and innovative ideas, and connect with international professionals working to end poverty on a livable planet. The Financial Engineering unit (ITSFE) within ITS supports the World Bank Group's core financial operations by designing and maintaining data pipelines, reporting systems, and analytical tools that underpin critical financial instruments — including IDA replenishments and disbursements. IDA, the World Bank's fund for the world's poorest countries, operates at massive scale and with the highest standards of data integrity. Any error or anomaly in the underlying data flows can cascade into financial reports relied upon by internal stakeholders, donor governments, and partner institutions. Traditional data engineering in this space relies on static, rule-based validation logic — an approach that is increasingly insufficient in the face of complex, high-volume, and evolving data environments. Machine learning offers a pathway to dynamic, adaptive data quality controls that can detect anomalies, flag missing data, and identify forecasting inconsistencies before they reach downstream systems. ITSFE is seeking a Pioneer intern to help design and prototype a machine learning–based anomaly detection capability integrated directly into IDA's data pipelines. This role sits at the intersection of data engineering, financial operations, and applied AI — offering a rare opportunity to contribute to global development finance through cutting-edge technology.

Requirements

Currently enrolled in, or in the final year of postgraduate program in Engineering.
0–6 years of relevant professional experience
Academic background must align with the requirements outlined in the job description
Strong statistical background, including understanding of probability distributions, time-series analysis, and anomaly detection methodologies.
Hands-on experience with machine learning libraries such as scikit-learn, TensorFlow, or PyTorch.
Proficiency in Python and data manipulation tools (pandas, NumPy, SQL).
Familiarity with cloud-based data engineering concepts, preferably on Azure.
Intellectually curious with a genuine interest in applying AI to high-impact, real-world financial systems.
Demonstrated interest in development work and the World Bank Group’s mission
Strong analytical, research, and problem-solving skills

Responsibilities

Apply machine learning algorithms to data pipelines handling IDA replenishments and disbursements to automatically flag anomalies, missing data patterns, and forecasting errors before they propagate into downstream financial reports.
Conduct a structured analysis of historical IDA data flows, including replenishment cycles, disbursement patterns, and associated metadata, to identify key signals and failure modes relevant to anomaly detection.
Design and train a lightweight, interpretable anomaly detection model using appropriate machine learning approaches (e.g., Isolation Forest, Autoencoders, or statistical process control methods), calibrated to the sensitivity requirements of financial data.
Document model assumptions, feature engineering decisions, and evaluation metrics in a clear and reproducible manner.
Integrate the trained model into an automated data pipeline leveraging Azure cloud services (e.g., Azure Data Factory, Azure Machine Learning, or Azure Databricks), in alignment with ITSFE's existing infrastructure.
Develop alerting or flagging mechanisms that surface detected anomalies to data engineers and financial analysts in a timely and actionable format.
Ensure the solution adheres to WBG data governance standards and security protocols.
Participate fully in ITSFE's Agile ceremonies, including sprint planning, daily standups, sprint reviews, and retrospectives.
Present progress and prototype demos to unit stakeholders, showcasing how predictive capabilities improve data governance and reduce manual validation overhead.
Collaborate with data engineers, financial analysts, and technical leads to refine requirements and validate model outputs against real-world expectations.
Produce technical documentation covering the model architecture, pipeline integration design, and operational guidelines for handoff to the engineering team.
Prepare a final presentation summarizing findings, methodology, and recommendations for scaling or productionizing the solution.