Junior Researcher AI for Youth Interventions Research

Uva; Amsterdam; Hybrid; EUR 42,000 - 54,000; Full time

Job summary

University of Amsterdam is seeking a Junior Researcher to design and implement a GenAI-assisted data extraction pipeline for SCED studies, and to build a Shiny application to inspect extracted data alongside figures.

You will work with clinical researchers, an information scientist, a software engineer and a student assistant, and contribute to open-access code and a scientific article.

Qualifications

• Master's degree required.
• Strong critical thinking and data-science affinity.
• Experience coordinating research projects in a team.
• Ability to translate technical plans into concrete steps.
• Experience with R or Python and working with large language models.
• Ask critical questions about AI output and document clearly for scientific audiences.

Responsibilities

• Design and implement GenAI-assisted data extraction pipeline for SCED studies.
• Develop prompting strategies to reconstruct symptom trajectories from graphs.
• Benchmark GenAI outputs against a manually coded reference dataset.
• Build an interactive Shiny application for data inspection and validation.
• Present findings at conferences and contribute to open-access papers.

Skills

Critical thinking

Team coordination

Communication

Translate complex plan

R/Python experience

AI QA mindset

Education

Master's degree

Tools

R

Python

Job description

Building a methodologically sound pipeline from raw data to treatment policy

You will coordinate and (co-)develop the design, implementation, and execution of the GenAI-assisted data extraction pipeline, including coding structure, benchmarking GenAI outputs against the reference data set, and integration into a Shiny app. This pipeline uses large language models to automatically extract case characteristics and information on symptom changes from single-case experimental design (SCED) studies. In SCEDs, treatment effectiveness is studied on a level of a single client using repeated observations of symptom change.

You will work closely with several investigators (clinical researcher, information scientist), a software engineer and a student assistant. Next to building a pipeline, you will share your findings with the research community through open-access code, a Shiny web application, a scientific article and a conference presentation.

What are you going to do

Under supervision, with input from the team and with a student assistant:

• Design and implement a GenAI-assisted pipeline to extract study and case characteristics from published PDF articles using large language models.
• Develop and refine prompting strategies to reconstruct symptom trajectories from graphs embedded in scientific publications.
• Benchmark automated GenAI outputs against a high-quality manually coded reference dataset to assess accuracy and identify error patterns.
• Build an interactive Shiny application that allows researchers to inspect and validate extracted data alongside original figures.
• Present findings at a conference and contribute to an open-access paper.

What do you have to offer

• You have a master-level degree.
• You have a critical thinking level, preferably with affinity for data science, research methodology, clinical psychology, or a related field.
• You have affinity with coordinating research projects and collaborating in a team or you would like to further develop your skills in these areas.
• You can translate, under supervision and in team collaboration, a complex technical plan into concrete steps.
• You have experience with R or Python and feel comfortable working with large language models or you would like to develop your skills further in these areas.
• You ask critical questions about AI output rather than taking results at face value.
• You communicate clearly — with your colleagues, the student assistant, and in writing for a scientific audience.

What else do we offer you

We offer a stimulating and interdisciplinary working environment, with a diverse range of tasks within an inspiring organization. Working at the University of Amsterdam means working in a dynamic, independent, creative, innovative, and international environment, characterized by an open atmosphere.

In this department, you will work

The Research Institute of Child Development and Education (RICDE) is part of the Faculty of Social and Behavioural Sciences at the University of Amsterdam. Dedicated to advancing knowledge on the development, learning, and well-being of children, the institute integrates cutting-edge research with real-world applications to impact education and care. We offer a dynamic environment paired with an informal and supportive atmosphere. You will work in an enthusiastic and warm team where initiative and input are highly valued.

The junior researcher will become part of MetaSCED lab. The lab is embeded in the Clinical Child and Family Studies group at the Research Institute of Child Development and Education (RICDE). Check out the program group via https://cde.uva.nl/research/child-development/clinical-child-and-family-studies/clinical-child-and-family-studies.html

Can AI free up clinical researchers by assisting them with tedious information extraction? For a period of ten months, we are looking for a Junior Researcher to test whether GenAI can reliably extract individual youth treatment data from scientific publications — accelerating evidence synthesis that helps professionals decide what works for which child. Join this one-year project at the cutting edge of AI and clinical research methodology.

Building a methodologically sound pipeline from raw data to treatment policy

You will coordinate and (co-)develop the design, implementation, and execution of the GenAI-assisted data extraction pipeline, including coding structure, benchmarking GenAI outputs against the reference data set, and integration into a Shiny app. This pipeline uses large language models to automatically extract case characteristics and information on symptom changes from single-case experimental design (SCED) studies. In SCEDs, treatment effectiveness is studied on a level of a single client using repeated observations of symptom change.

You will work closely with several investigators (clinical researcher, information scientist), a software engineer and a student assistant. Next to building a pipeline, you will share your findings with the research community through open-access code, a Shiny web application, a scientific article and a conference presentation.

What are you going to do

Under supervision, with input from the team and with a student assistant:

• Design and implement a GenAI-assisted pipeline to extract study and case characteristics from published PDF articles using large language models.
• Develop and refine prompting strategies to reconstruct symptom trajectories from graphs embedded in scientific publications.
• Benchmark automated GenAI outputs against a high-quality manually coded reference dataset to assess accuracy and identify error patterns.
• Build an interactive Shiny application that allows researchers to inspect and validate extracted data alongside original figures.
• Present findings at a conference and contribute to an open-access paper.

What do you have to offer

• You have a master-level degree.
• You have a critical thinking level, preferably with affinity for data science, research methodology, clinical psychology, or a related field.
• You have affinity with coordinating research projects and collaborating in a team or you would like to further develop your skills in these areas.
• You can translate, under supervision and in team collaboration, a complex technical plan into concrete steps.
• You have experience with R or Python and feel comfortable working with large language models or you would like to develop your skills further in these areas.
• You ask critical questions about AI output rather than taking results at face value.
• You communicate clearly — with your colleagues, the student assistant, and in writing for a scientific audience.

What else do we offer you

We offer a stimulating and interdisciplinary working environment, with a diverse range of tasks within an inspiring organization. Working at the University of Amsterdam means working in a dynamic, independent, creative, innovative, and international environment, characterized by an open atmosphere.

In this department, you will work

The Research Institute of Child Development and Education (RICDE) is part of the Faculty of Social and Behavioural Sciences at the University of Amsterdam. Dedicated to advancing knowledge on the development, learning, and well-being of children, the institute integrates cutting-edge research with real-world applications to impact education and care. We offer a dynamic environment paired with an informal and supportive atmosphere. You will work in an enthusiastic and warm team where initiative and input are highly valued.

The junior researcher will become part of MetaSCED lab. The lab is embeded in the Clinical Child and Family Studies group at the Research Institute of Child Development and Education (RICDE). Check out the program group via https://cde.uva.nl/research/child-development/clinical-child-and-family-studies/clinical-child-and-family-studies.html

For questions about the vacancy, you can contact: Marija Maric, [e-mail].