Postdoc in Data Science and Statistics

Om jobbet

Can satellites see poverty-and can AI help end it? We are looking for a Postdoctoral Researcher to join a pioneering effort at the intersection of statistics, data science, earth observation, and global development. In this position, you will develop methods to estimate poverty across the African continent using satellite imagery, contribute to open-source tools for policy evaluation, and advance causal inference techniques that can reveal whether development interventions actually work. You will be part of an international, interdisciplinary team with collaborators at Harvard University, the University of Texas at Austin, and leading Swedish universities-and you will be mentored by scholars who are committed to your growth as a researcher.

About us

The Department of Computer Science and Engineering, a joint department of Chalmers and the University of Gothenburg, spans the breadth of computing disciplines. Our internationally visible research, strong industry links and diverse environment create a collaborative setting where ideas grow into real impact.

At the division of Data Science and AI, we develop data-driven methods and AI solutions that support intelligent decisions across society, advancing machine learning techniques, from foundations to industrial and scientific applications.

About the Lab

This position is in the Data Science and AI (DSAI) division at the Department of Computer Science and Engineering, Chalmers University of Technology. The postdoc will join the AI and Global Development Lab (www.aidevlab.org), funded by the Swedish Research Council (VR), which furthers the use of AI for Social Good in pursuit of the sustainable development goals.

The Lab brings together collaborators based in Sweden, the United States, India, Chile, and the United Kingdom, and publishes in top interdisciplinary generalist journals, discipline-specific journals, and conferences. As the project is highly interdisciplinary, we welcome applicants from a variety of disciplinary backgrounds and will adapt our publication strategy depending on the candidate's background and interest.

Work environment

The Lab meets weekly both remotely and in-person, with collaborators across multiple time zones. The project constitutes a collaboration mainly among the Department of Computer Science and Engineering at Chalmers, the Department of Political Science at the University of Gothenburg, the Institute for Analytical Sociology at Linköping University (campus Norrköping), and the Department of Statistics at Harvard University. Occasional travel within Sweden and abroad is expected.

Leadership and mentorship

The Lab is headed by Adel Daoud (www.adeldaoud.com), who will serve as the primary mentor. Daoud is a computational social scientist, Professor in Computational Social Science at the Institute for Analytical Sociology (Linköping University), and Affiliated Associate Professor in Data Science and AI for the Social Sciences at Chalmers. He has previously held positions at Harvard University, the University of Cambridge, and the Alan Turing Institute.

Other mentors include senior Lab members and collaborators such as Connor Jerzak at the University of Texas at Austin (www.connorjerzak.com) and Mohammad Kakooei at Karlstad University, as well as Devdatt Dubhashi (DSAI) and Xiao-Li Meng (Harvard).

The Lab is committed to providing high-quality mentorship. The candidate is encouraged to explore the Journeys of Scholars podcast (created by Daoud), featuring conversations about the trajectories, strategies, and advice of leading academic scholars-available on YouTube and Spotify. See https://www.youtube.com/@thejourneysofscholars8820

About the Research Project

About 900 million people-one-third in Africa-live in extreme poverty. Scholars are currently unable to determine to what extent poverty traps exist and whether development interventions release communities from such traps, because geo-temporal poverty data are scarce.

The Observatory of Poverty project, funded by the Swedish Research Council, addresses this challenge with two main goals:

Develop new methods to produce geo-temporal poverty data by training deep learning algorithms to estimate poverty from satellite images of African communities, quarterly, from 1984 to 2025; and contribute to creating the ObservatoryOfPoverty statistical package that enables scholars to produce poverty estimates for policy evaluation.

Use these estimates, combined with causal inference methods, to evaluate how competing aid and development interventions alter communities' prospects of escaping deprivation.

As part of this effort, the project also develops prediction-assisted estimation and inference methods-including prediction-powered inference and conformal prediction-to ensure that satellite-derived estimates support valid statistical reasoning for policy evaluation.

More information: www.aidevlab.org

What you will do

Train deep learning algorithms to estimate poverty from satellite images of African communities over time and space (quarterly, 1984-2025).

Contribute to creating the ObservatoryOfPoverty statistical package, enabling scholars to produce poverty estimates for policy evaluation.

Advance methodological development in causal inference and prediction-assisted estimation and inference for the project, including prediction-powered inference, conformal prediction, and multiple imputation for missing data.

Publish research in top interdisciplinary generalist journals, discipline-specific journals, and conferences.

Supervise master's and/or PhD students to a certain extent.

Possibility to engage in teaching at undergraduate/master's level (up to 20% of time).

The position provides strong preparation for future roles in academia, industry, or the public sector.