ML Engineer

KEMIO Consulting; Greater London; Hybrid; GBP 90,000 - 130,000; Full time

Job summary

KEMIO Consulting is partnering with a biotechnology company in London to hire an engineer who will turn cutting-edge research into robust, scalable production systems for biomedical AI models. You will work with AI Scientists from model development through production deployment, ensuring research ideas become reliable software used across the organization.

You'll help design and build large-scale models and production pipelines, collaborating with researchers and MLOps to deliver scalable,

Qualifications

• PhD in Machine Learning, Computer Science, Computational Biology or similar.
• 3–6 years post-study work experience with biomedical datasets (omics, clinical, imaging).
• Strong experience developing deep learning models and foundation model architectures (transformers, pre-training, fine-tuning).
• Excellent Python programming skills and experience with PyTorch or JAX.
• Experience taking ML research from prototype to production deployment.
• Strong software engineering fundamentals: testing, documentation, code reviews, version control.

Responsibilities

• Partner with AI Scientists to transform research into production-ready ML systems.
• Contribute to architecture and implementation of large-scale foundation models for efficiency and deployment-readiness.
• Develop training pipelines, data loaders, tokenisation frameworks, inference services and fine-tuning workflows.
• Build clean, maintainable and well-tested Python code following software engineering practices.
• Benchmark and optimize model performance and training efficiency at scale.
• Collaborate with MLOps to ensure smooth deployment and ongoing maintenance; document capabilities and retraining strategies.
• Stay up to date with advances in ML engineering, distributed training and biomedical AI.

Skills

Python

Software engineering

Distributed training

Deep learning

Education

PhD in ML, CS, CompBio

3–6 years biomedical data experience

Tools

PyTorch

JAX

DeepSpeed

Ray Train

FSDP

Job description

London - Hybrid (3 days a week in the office)

We are partnering with an innovative, research-driven biotechnology company that is building next generation AI systems that help identify new therapeutic opportunities and accelerate the drug discovery process.

This is a hands-on engineering role for someone who enjoys turning state-of-the-art research into robust, scalable production systems. You'll work closely with AI Scientists from the earliest stages of model development, ensuring that research ideas become reliable, high-performance software used throughout the organisation.

You'll contribute to the design, development and production of large-scale biomedical AI models, bringing engineering expertise into architectural decisions from day one. Working alongside researchers and MLOps engineers, you'll help build production ready machine learning systems that are scalable, maintainable and built to the highest engineering standards.

Key Responsibilities

• Partner with AI Scientists to transform validated research into production-ready machine learning systems.
• Contribute to the architecture and implementation of large-scale foundation models, ensuring they are efficient, scalable and deployment-ready.
• Develop high-quality training pipelines, data loaders, tokenisation frameworks, inference services and fine-tuning workflows.
• Build clean, maintainable and thoroughly tested Python code following software engineering best practices.
• Benchmark and evaluate model performance while helping optimise training efficiency and scalability.
• Collaborate closely with MLOps teams to ensure smooth deployment, documentation and ongoing model maintenance.
• Produce comprehensive technical documentation covering model capabilities, limitations and retraining strategies.
• Stay up to date with emerging developments in machine learning engineering, distributed training and biomedical AI.

Requirements;

• A PhD in Machine Learning, Computer Science, Computational Biology or similar. Plus 3-6 years of post study work experience, working with biomedical datasets such as omics, clinical or imaging data.
• Must have strong experience developing deep learning models and foundation model architectures, including transformers, pre-training and fine-tuning.
• Extensive experience taking machine learning research from prototype through to production-quality deployment.
• Excellent Python programming skills and experience with frameworks eg PyTorch or JAX.
• Strong software engineering fundamentals, including testing, documentation, code reviews and version control.
• Experience with distributed training technologies such as PyTorch Distributed, DeepSpeed, FSDP or Ray Train.
• Experience working alongside research scientists to deliver production machine learning systems.