Multimodal AI Researcher (Intern)

We’re searching for a Multimodal AI Researcher (intern) to join out client on an initial 6 month contract working on a hybrid role in Surrey. 

The role is working with a globally recognised tech/consumer electronics company as as this is an internship, would suit a soon to be or fresh PhD graduate.

As a Multimodal AI Intern, you will:

• Drive innovation by developing cutting-edge solutions to real-world problems in on-device audio-visual AI.

• Proactively propose and prototype novel research ideas, factoring in practical constraints.

• Contribute to the development of complex AI systems, transforming research concepts into production-ready software.

• Apply software engineering best practices across both research and development phases.

• Effectively communicate research findings through technical reports and/or academic publications.

Required skill

• Recent graduate or currently pursuing a Ph.D. in Machine Learning/AI, Computer Science/Engineering, Mathematics, Statistics, or a related field.

• Strong foundational knowledge in machine learning and artificial intelligence.

• First-author publications in top-tier ML/AI conferences or journals (e.g., CVPR, ICCV, NeurIPS, ICML, ICLR, ICASSP, INTERSPEECH, IEEE TPAMI, IEEE IoT, IEEE TNNLS, JMLR, or similar).

• Hands-on machine learning experience in at least one of the following areas:

  • Multimodal LLMs (audio and/or video)

  • Contrastive learning (e.g., multimodal feature alignment)

  • Model compression techniques (e.g., quantization, pruning, knowledge distillation)

• Proven success in:

  • Software development using Python and/or C/C++

  • Working with ML frameworks such as PyTorch and/or TensorFlow

  • Writing clear and thorough documentation

  • Using standard software engineering tools and practices (e.g., Git)

• Strong communication and collaboration skills with a results-driven mindset

• Excellent problem-solving and debugging abilities

Preferred skills

• Background in multimodal emotion recognition and foundational facial models

• Experience with multi-task learning and deception detection

• Demonstrated ability to build sophisticated training and inference pipelines

• Knowledge of embedded and/or distributed ML tools and methodologies

• Experience in optimizing and profiling AI pipelines for performance

• Contributions to open-source machine learning libraries