Computer Vision Scientist | Multimodal AI / Vision-Language Models / Deep Learning

European Tech Recruit; Greater London; On-site; GBP 90,000 - 120,000; Full time

Job summary

European Tech Recruit is partnering with a London-based leader in AI research to hire a Research Scientist in Computer Vision and multimodal AI. The role focuses on next‑generation vision-language models, large-scale multimodal systems, and production-ready AI applications.

You will explore ViT architectures, cross-modal understanding, and scalable training across distributed GPUs, while curating huge multimodal datasets and building robust data pipelines.

Qualifications

• Bachelor's degree in a technical discipline and strong research background in computer vision or multimodal AI.
• Solid Python and PyTorch experience with modern deep learning frameworks.
• Strong understanding of ML algorithms, models, optimization, and mathematics.

Responsibilities

• Research and develop next-generation ViT, multimodal, and vision-language architectures with improved reasoning, efficiency, and generalisation.
• Advance multimodal representation learning, alignment, long-context modelling, and cross-modal understanding.
• Develop scalable training methodologies for large multimodal models across distributed GPU infrastructure.
• Explore new model architectures, training strategies, and optimisation techniques to improve model performance and efficiency.
• Process and curate large-scale multimodal datasets spanning images, video, audio, and text.
• Design data pipelines for dataset cleaning, filtering, annotation, quality control, versioning, and reproducibility.
• Develop data mixture and sampling strategies to improve training efficiency and model capabilities.

Skills

Python

PyTorch

Deep Learning

Vision-Language Models

Research Experience

Education

Bachelor's degree or higher in CS/Math/EE

Job description

Computer Vision Scientist | Multimodal AI / Vision-Language Models / Deep Learning

We are currently partnered with a leading global technology company with a major research and development presence in London, focused on advancing the frontiers of artificial intelligence, computer vision, multimodal understanding, and embodied intelligence.

As part of their continued investment in fundamental and applied AI research, they are looking to hire a Research Scientist specialising in Computer Vision and multimodal AI to contribute to next-generation vision-language models, large-scale multimodal learning systems, and production-ready AI applications.

Key responsibilities

• Research and develop next-generation Vision Transformer (ViT), multimodal, and vision-language model architectures with improved reasoning, efficiency, and generalisation.
• Advance multimodal representation learning, alignment, long-context modelling, and cross-modal understanding.
• Develop scalable training methodologies for large multimodal models across distributed GPU infrastructure.
• Explore new model architectures, training strategies, and optimisation techniques to improve model performance and efficiency.
• Process and curate large-scale multimodal datasets spanning images, video, audio, and text.
• Design data pipelines for dataset cleaning, filtering, annotation, quality control, versioning, and reproducibility.
• Develop data mixture and sampling strategies to improve training efficiency and model capabilities.

Key requirements

• Bachelor's degree or higher in Computer Science, Mathematics, Statistics, Electrical Engineering, or a related technical discipline.
• Strong commercial or academic research experience in computer vision, multimodal AI, deep learning, or related fields.
• Strong Python programming skills with substantial hands‑on experience using PyTorch and modern deep learning frameworks.
• Strong understanding of machine learning algorithms, deep learning architectures, optimisation techniques, and mathematical foundations.
• Experience designing, implementing, training, and evaluating sophisticated AI models.
• Strong publication record at leading AI and computer vision conferences including CVPR, ICCV, ECCV, NeurIPS, ICML, or ICLR.
• Hands‑on experience with large‑scale foundation model pre‑training or fine‑tuning.
• Experience developing Vision-Language Models (VLMs), Multimodal Large Language Models (MLLMs), or other multimodal foundation models.

Keywords

Research Scientist / Computer Vision Scientist / AI Research Scientist / Machine Learning Researcher / Computer Vision Researcher / Multimodal AI / Multimodal Machine Learning / Vision-Language Models / VLM / Multimodal Large Language Models / MLLM / Large Language Models / LLM / Vision Transformer / ViT / Deep Learning / PyTorch / Python / Foundation Models / Generative AI / Representation Learning / Multimodal Alignment / Long Context / Computer Vision / Image Understanding / Video Understanding / AI Research / Distributed Training / GPU Computing / Model Pre-Training / Fine-Tuning / Model Serving / AI Infrastructure / Data Curation / Dataset Development / CVPR / ICCV / ECCV / NeurIPS / ICML / ICLR / London / UK AI Research

By applying to this role, you understand that we may collect your personal data and process it in line with our privacy policy.