Opportunities

Research Fellow — DMFP Project

The DMFP (Diffusion Modelling for Print) project is developing a generative modelling pipeline for industrial and consumer design applications. Our research explores diffusion and GAN architectures, raster-to-vector conversion techniques, and hybrid image representations — with the goal of creating physically-constrained image synthesis systems suitable for professional design workflows.

Preferred Background

We welcome applicants with backgrounds in Computer Science, Machine Learning, Artificial Intelligence, or Mathematics. Advanced undergraduates (final year) and postgraduate students are preferred. Candidates should have foundational knowledge of machine learning theory. Experience with image generation, conversion, or compression algorithms is advantageous but not required. Proficiency in Python with PyTorch or TensorFlow is beneficial — we provide mentorship for skill development.

While the research area may seem specialised, our team is committed to supporting your learning journey. We encourage applications from motivated candidates eager to explore this field.

Time Commitment

Approximately 8–14 hours per week, including one weekly team meeting and 1–2 additional synchronisation calls as required.

What You'll Gain

• Practical experience with state-of-the-art generative models
• Exposure to industry-aligned research challenges
• Collaborative reading groups and technical discussions
• Potential co-authorship on publications
• Portfolio-quality research contributions