AI Scientist / Machine Learning Engineer

Overview

Efestos Hub is growing its AI capability to support Innovate UK-backed research and development work in circular construction, structural steel reuse, and automated assessment of reclaimed materials. The team is seeking three AI Scientists / Machine Learning Engineers to build applied machine learning and computer vision solutions for the built environment.

This position sits at the crossroads of artificial intelligence, structural engineering, circular economy, construction technology, low-carbon material reuse, and computer vision applied to real industrial data.

The core challenge is to create AI systems that can analyse images and structured information from reclaimed steel components, detect visible condition characteristics, and help guide decisions around reuse, processing, re-fabrication, and recovery.

What you will work on

You will contribute to machine learning and computer vision development for practical steel reuse use cases, including image-led condition assessment of reclaimed steel sections, defect detection, and AI outputs that support engineering decisions.

• Assessing the condition of reclaimed steel sections from images
• Identifying and classifying defects such as corrosion, holes, attachments, dents, weld scars, and surface deterioration
• Building CNN and deep learning solutions for image classification, object detection, and segmentation
• Connecting visual findings to engineering and process decisions
• Creating explainable outputs such as confidence scores and reason codes
• Preparing datasets, shaping labelling approaches, carrying out quality checks, and evaluating models
• Working with engineers and industry partners to verify model performance
• Helping integrate AI models into Efestos' digital platform

Skills and experience

Strong Python and applied AI experience are essential, along with hands-on work in machine learning, deep learning, and computer vision on imperfect real-world datasets.

• Python programming
• Machine learning and deep learning
• Convolutional neural networks
• PyTorch and/or TensorFlow/Keras
• Computer vision with OpenCV or similar libraries
• Working with messy, real-world image datasets
• Classification, object detection, and/or segmentation workflows
• Model training, validation, testing, and performance reporting
• Clear coding practices, documentation, and Git-based collaboration

Desirable experience

• Experience with YOLO, Mask R-CNN, U-Net, Detectron2, Segment Anything, or comparable models
• Use of annotation tools such as CVAT, Label Studio, or Roboflow
• Exposure to MLOps, experiment tracking, or model deployment
• Familiarity with FastAPI, Docker, cloud/GPU environments, or model APIs
• Background in engineering, construction, materials, manufacturing, inspection, or robotics
• Interest in sustainability, circular economy, and low-carbon construction

Candidate background

Efestos Hub welcomes applicants from a range of technical backgrounds, including AI/ML engineers, computer vision researchers, PhD students, postdoctoral researchers, data scientists with image-analysis experience, software engineers with applied ML experience, and civil, structural, or mechanical engineers with strong AI capability.

No prior steel reuse experience is required, but candidates should be motivated by applying AI to a real industrial problem with measurable environmental impact.

Contract details

This is a fixed-term contract role tied to funded R&D delivery. The company is looking for someone who can begin quickly and contribute over the next few months in a fast-moving technical programme.

The role can be part-time or full-time depending on availability, experience, and project fit.

About Efestos Hub

Efestos Hub is building AI-enabled tools that support the reuse of structural steel and circular construction. Its platform turns buildings, stockyards, and reclaimed components into digital inventories to enable lower-carbon procurement and large-scale reuse.

The company describes its work as building the intelligence layer for the future steel reuse market.

How to apply

Applicants should send a CV and, where available, GitHub links, portfolios, publications, or examples of previous work, along with availability and preferred working arrangement. The email subject line requested is Efestos-July26-AI Scientist.

The company is especially interested in candidates who can combine strong technical depth with practical delivery.

Additional information

Location preference is for UK-based candidates, with remote or hybrid working available. The role is intended to start as soon as possible.

Role count

Three positions are being recruited for this AI Scientist / Machine Learning Engineer hiring round.