AI4ME is a project in partnership with the BBC and the University of Surrey developing next-generation methods for the delivery of multimedia.
As part of this project we are looking for a research technician to support the implementation of content avoidance methods in a media delivery pipeline. This will require the use of machine learning and AI techniques.
Content avoidance is the removal of video footage that a viewer does not want to see. This could be a spider for an arachnophobe, or something graphic such as blood and gore. With advancements in AI and Visual Reasoning Models it may also be something the user can personalise themselves, describing the footage they do not wish to see.
The expected deliverables for this role include:
- Tooling for automated content avoidance, e.g. removing spiders by blurring them out.
- Developing an understanding of the performance of different tools in terms of speed and accuracy.
- Integrating content avoidance tooling into an existing Web Assembly Media Player.
- Demos for showcasing at project events.
The ideal candidate will have a strong background in computer science, and proficient coding skills. It would also be desirable to have an understanding of delivery of video content, and have an interest in applying ML/AI techniques to video. A good candidate would also be independent, able to present findings back to both academic and industry stakeholders, and have a desire to build tools or systems that can be demonstrated to the BBC and other stakeholders.
This role is advertised on a part time basis, with the aim to complete at most 16 hours per week. The work can be remote or in-person. This is flexible, with the expectation that the successful candidate will have other responsibilities they need to fit around this role (e.g. studying).
Preferred start date: 7th of September 2026
Expected end date: 30th of November 2026
Interview date: 28th of August 2026
Weekly hours: 16 hours per week
Location: Remotely (in the UK) and on the Lancaster University campus