Student Assistant in AI for Spotting Toxic Clouds

The Faculty of Science at the University of Amsterdam (UvA) has an opening for a student assistant in AI for Spotting Toxic Clouds to help program a software tool for a citizen science project.

The tool will be used by local communities in the IJmond region of the Netherlands to monitor air pollution from a steel-making factory using camera data. IJmondCAM aims to train an artificial intelligence model to spot toxic clouds, and they need volunteers to help label smoke emissions. So far, 845 (99.76%) videos are fully labeled. Their goal is to invite citizens to create data-driven evidence of pollution events (using the smoke emissions found by the AI model). They hope that this evidence can facilitate air quality advocacy in the IJmond region of the Netherlands.


More specifically, student tasks will include:

  • Build a data engineering pipeline to extract video clips from the camera stream for citizens to label whether the clips contain smoke emissions or not, based on available code from the AI against Toxic Clouds challenge;
  • Adapt an available computer vision pipeline to detect smoke emission in the IJmond camera data, which includes running experiments to evaluate model performance, and also deploying the model to process the live camera stream;
  • Engineer a dashboard for local communities to better understand smoke emissions from the steel factory (e.g., how bad the emission is, how frequent the emissions are, and how likely the emissions will violate air pollution law).

Duration: 7 months.

Starting Date: February 2024.


  • The salary depends on the academic year: €2,618 (2nd year bachelor’s) to €2,846 (master’s) gross per month based on full-time working time (38 hours per week).
  • UvA additionally offers an extensive package of secondary benefits, including an 8% holiday allowance and a year-end bonus of 8.3%.
READ ALSO  PhD Position in “Water Footprint Assessment and Water Allocation Modelling”

Job Requirements

Applicants for this position must be registered students at the University of Amsterdam.

Application Process

Applications should include the following information (all files apart from CVs should be submitted in one single pdf file):

  • A detailed CV, including the months (not just years) when referring to education and work experience;
  • A letter of motivation.

To apply for the Student Assistant in AI for Spotting Toxic Cloudsclick here.

Deadline: March 1, 2024.

For more information on Student Assistant in AI for Spotting Toxic Clouds, visit the official site.

Verified by MonsterInsights