(Contract) Computer Vision / Machine Learning Engineer

🕙 Role details: 2 month contract (Mid Oct - Mid December)

💷 Day rate: TBC / Outside IR35

The role

(Please note, this is a fixed-term contract, not a permanent position)

Make your mark in digital healthcare

Isla is transforming the way healthcare is delivered and we’re looking for a Computer Vision / AI Engineer (Contractor) to support an NIHR-funded programme called “WISDOM”, focused on improving post-operative wound monitoring through artificial intelligence.

This short-term contract (mid-October to mid-December) will play a critical role in strengthening our AI wound analysis module. You will be advising on image preprocessing & annotation, and leading on model development and validation across diverse patient populations.

To succeed, you must have:

• Proven expertise in computer vision and deep learning, ideally in medical imaging
• Experience with data preprocessing and annotation pipelines
• A track record of developing, training, and validating AI models with real-world datasets
• Familiarity with evaluating model performance across subgroups (e.g. skin tones, wound types)
• Awareness of health data standards, GDPR, and clinical compliance requirements

---

What is Isla?

Isla is the award-winning digital pathway platform transforming the way healthcare is delivered. Built with clinicians, for clinicians, it turns repetitive, manual tasks into intelligent, scalable workflows freeing capacity, accelerating decisions, and improving patient outcomes.

Already live in 30+ NHS Trusts and across 40+ specialties, Isla securely captures photos, videos, sound recordings and questionnaires from patients and clinicians. This data powers automated, personalised pathways and clinically coded risk stratification, giving teams complete visibility from triage to recovery.

Proven results at scale:

• 85% faster referral-to-treatment times
• 3.8× more clinic capacity
• 300–400% ROI guaranteed within three months

With 2M+ secure submissions to date (one every three minutes) Isla is delivering £3M+ potential annual savings per Trust, cutting 3.6M+ patient travel miles, and enabling 30% of patients to be moved to remote care.

---

What is Wisdom?

Surgical wound infections are a significant challenge, affecting 5% of patients and costing the NHS an estimated £1 billion each year. While early detection is key to reducing both the severity of these infections and their treatment costs, widespread monitoring isn't currently happening.

Digital monitoring platforms, where patients can submit photos of their wounds, offer a promising solution but can increase the workload for clinicians. Our initial study developed a well-received AI platform to address this, but we now need to take it a step further.

Our primary goal is to enhance the AI model's accuracy. Specifically, we need to improve its sensitivity for identifying infections, and critically, its ability to detect early infection signs darker skin tones. This project is a chance to make a real impact on patient care and health outcomes.

---
What you’ll be doing

Over the course of the contract you will:

• Conduct interim sensitivity analyses against ground-truth labels
• Develop and retrain AI models to improve detection of infection and wound healing
• Explore and integrate state-of-the-art public models (e.g. Gemini), evaluating opportunities for fine-tuning, hybrid approaches, and performance benchmarking
• Test and validate model equity performance across different patient groups, including light and dark skin tones
• Document workflows and ensure compliance with NHS software standards and GDPR
• Collaborate closely with clinicians, researchers, and product teams to align technical outputs with clinical needs
• Contribute to reporting for NIHR and project stakeholders

---

Requirements

• Strong experience in computer vision / deep learning (e.g. CNNs, segmentation models, classification pipelines)
• Expertise in Python and relevant ML frameworks (e.g. PyTorch, TensorFlow, Keras)
• Experience handling large and diverse image datasets
• Solid understanding of sensitivity/specificity metrics, ROC analysis, and subgroup validation
• Awareness of bias and fairness considerations in AI models
• Experience in healthcare AI or medical imaging projects (desirable)
• Knowledge of NHS digital standards, GDPR, or medical device compliance (nice to have)