Mott Macdonald
A global engineering, management, and development consultancy delivering innovative solutions.
Data Scientist
Develop and deploy advanced computer vision and multimodal AI solutions for engineering.
Job Highlights
About the Role
The Data Scientist (Computer Vision) will lead the development of intelligent systems that turn satellite imagery, site photos, and video into actionable engineering insights, integrating these capabilities into the company's digital‑twin platform. • Implement state‑of‑the‑art Vision Transformers and multimodal VLLMs for complex scene understanding. • Design automated systems for detecting infrastructure defects, safety hazards, and environmental changes. • Build and deploy vision‑language models (e.g., LLaVA, CLIP) enabling natural‑language interaction with visual data. • Develop high‑performance serving architectures using vLLM and Azure AI to scale inference of high‑resolution imagery and video.
Key Responsibilities
- ▸vision transformers
- ▸defect detection
- ▸model deployment
- ▸serving architecture
- ▸satellite imagery
- ▸digital twin
What You Bring
• Own end‑to‑end lifecycle of computer‑vision projects, from data acquisition and model selection to deployment on the digital‑twin platform. • 5+ years of experience in computer vision and deep learning with strong mathematical foundations. • Hands‑on expertise with Vision Transformers (ViT, Swin, DeiT) and detection/segmentation models such as DETR, Mask2Former, SAM. • Proven work with Vision Large Language Models (LLaVA, Phi‑Vision, BLIP) for visual question answering and captioning. • Proficiency in PyTorch (or TensorFlow) and Python, emphasizing clean, scalable code. • Experience deploying vision pipelines on Azure ML, including managed endpoints, MLOps, and high‑resolution dataset handling. • Familiarity with inference optimization techniques (vLLM, quantization methods like AWQ or GPTQ) is a plus. • Passion for applying AI to engineering, geospatial science, or urban planning. • Ability to communicate complex multimodal AI concepts to non‑technical stakeholders. • Innovative “hacker” mindset, staying abreast of CVPR/NeurIPS research and translating state‑of‑the‑art papers into production. • Agile/remote working flexibility.
Requirements
- ▸vision transformers
- ▸pytorch
- ▸azure ml
- ▸mlops
- ▸5+ years
- ▸agile
Benefits
Mott MacDonald offers purpose‑driven work that addresses global challenges, continuous AI and leadership learning paths, an inclusive culture, and flexible working arrangements to support work‑life balance. Employees benefit from agile working, comprehensive insurance, family leave, career mobility, global collaboration, and both short‑ and long‑term employment opportunities. • Critical illness, compassionate, paternity, and other family leaves. • Group term life and medical insurance. • Career mobility and global employment opportunities. • Access to AI, data‑science, and leadership learning programs.
Work Environment
Hybrid