Machine Learning Engineer

Develop the Next Generation of AI for Computational Life Sciences

At JIJ, we develop advanced computational technologies for solving complex scientific and industrial problems. Our R&D spans machine learning, mathematical optimisation, network science, quantum and quantum-inspired computing, with a strong emphasis on translating fundamental research into practical scientific impact.

We are expanding our interdisciplinary life-sciences research team to develop advanced computational approaches for challenging problems in microbiology, bioinformatics, genetics, microbiome science, and biological network modelling. Example problems include understanding how microbial communities change over time, how microorganisms interact, how biological perturbations propagate through these communities, and which mechanisms determine recovery or loss of resilience.

This is not a role focused simply on applying standard machine-learning pipelines to biological datasets. It is particularly suited to researchers interested in developing new computational methodology and translating research ideas into reproducible scientific technology.

The Opportunity

We are looking for a Machine Learning Engineer with expertise spanning computational biology and machine learning to develop next-generation models of longitudinal microbiome systems.

You will work with high-dimensional biological datasets including microbial abundance, genomic and functional information, environmental or intervention metadata, and potentially metabolomic and other multi-omics measurements. A central research direction will be representing biological systems as dynamic graphs, where microorganisms, genes, metabolites, pathways, or host factors form interacting networks whose topology and state evolve over time.

You will investigate state-of-the-art approaches including:

• Graph machine learning: GNNs, message passing, graph attention, graph transformers, and heterogeneous or knowledge-graph models.
• Temporal and dynamic modelling: evolving interaction networks and continuous-time or state-space approaches.
• Biologically informed learning: combining pathway, phylogenetic, metabolic, and genomic priors with data-driven models.
• Representation and multi-omics learning: self-supervised methods, foundation-model representations, and multimodal integration.
• Causal, interpretable, and uncertainty-aware modelling for perturbation analysis, biological discovery, and robust prediction.

Where scientifically justified, you may also investigate quantum or quantum-inspired approaches (e.g. quantum machine learning, quantum kernels, tensor networks, Ising/QUBO formulations). These are complementary research directions rather than mandatory components of every project.

What You'll Do

Computational Biology & Biological Representation

• Develop computational models for microbiome dynamics, perturbation, recovery, resilience, and broader biological network problems.
• Build biologically meaningful graph and multi-omics representations that integrate microbial, genomic, metabolic, pathway, phylogenetic, and host information.
• Formulate biologically meaningful questions and translate them into rigorous computational experiments.

Graph, Temporal & Interpretable Learning

• Design and evaluate GNN, graph-transformer, temporal, causal, and interpretable learning methods.
• Address microbiome-specific challenges such as compositionality, sparsity, batch effects, confounding, and irregular sampling.
• Establish strong classical and statistical baselines and distinguish statistical associations from biologically plausible mechanisms.

Research to Impact

• Translate state-of-the-art research into reproducible experiments, research software, technical reports, and peer-reviewed publications.
• Collaborate across computational and experimental teams and with academic, scientific, and industry partners.
• Develop reusable methods and technologies rather than one-off analyses wherever possible.

Qualifications

We are looking for engineers and researchers who can move comfortably between biology, mathematics, and machine learning. You may be an excellent fit if you have:

• A PhD, or a research-intensive MSc with significant experience, in Computer Science, Machine Learning, Bioinformatics, Computational Biology, Microbiology, Genetics, Genomics, Systems Biology, or a closely related discipline.
• A strong background in statistics and machine learning, with the ability to design, train, validate, and rigorously benchmark predictive models for complex scientific datasets.
• Practical experience developing models in Python using PyTorch, JAX, TensorFlow, or an equivalent framework.
• Experience with graph machine learning or biological network modelling (GNNs, graph attention or transformers, heterogeneous graphs, knowledge graphs, or temporal graph models).
• Strong working knowledge of microbiology, microbial ecosystems, genetics, molecular biology, or bioinformatics, enabling effective communication with biological collaborators.
• Experience analysing sequencing or omics data such as 16S, metagenomics, microbial genomics, transcriptomics, or metabolomics.
• The ability to formulate biologically meaningful computational questions, design rigorous experiments and baselines, and communicate across computational and experimental disciplines.

Preferred Qualifications

Experience in one or more of the following would be particularly valuable:

• Computational biology: longitudinal microbiome, microbial ecology, or host–microbiome modelling; metabolic, phylogenetic, or multi-omics analysis; systems-biology modelling; compositional data analysis and microbiome statistics.
• Machine learning: dynamic GNNs or graph transformers; neural differential equations or state-space models; causal inference, perturbation modelling, explainable AI, or uncertainty quantification; biologically informed neural networks; self-supervised learning, genomic language models, or biological foundation models.
• Research & engineering: reproducible experimentation and benchmarking; academic or industrial publications; production-quality scientific software; open-source contributions; collaboration with experimental scientists.

Experience with tools such as QIIME 2, DADA2, Bioconductor, MetaPhlAn, PyTorch Geometric, or DGL, or with quantum and quantum-inspired methods, is welcome but not required. We also welcome candidates whose strongest expertise is not explicitly listed above.

What We Are Looking For

A strong candidate should be able to look at a microbiome dataset and ask questions such as:

• What constitutes a biologically meaningful edge between two microorganisms?
• Is an observed interaction ecological, metabolic, phylogenetic, causal, or merely statistical?
• How should interactions and graph structure evolve through time or under perturbation?
• Can prior biological knowledge improve generalisation and distinguish correlation from plausible mechanism?

We value candidates who challenge modelling assumptions, establish rigorous baselines, and develop methods that are both computationally sophisticated and biologically defensible.

Why Join JIJ?

• Develop new AI and graph-learning methodology at the intersection of biology, mathematics, and machine learning.
• Work with longitudinal, multi-omics, and multimodal biological datasets in high-impact collaborative research programmes.
• Collaborate with specialists in bioinformatics, machine learning, optimisation, network science, quantum computing, and experimental life sciences.
• Contribute to peer-reviewed publications, technical reports, and reproducible research software.
• Influence the technical direction of JIJ's growing computational life-sciences research activities.

Location & Working Arrangement

• Location: United Kingdom or EU
• Employment: Full-time
• Working style: Remote-first, with periodic travel to London for research collaboration, workshops, partner engagements, and team meetings. Flexible working hours.
• Visa sponsorship: Visa sponsorship and relocation support may be available for eligible candidates.

Compensation

Annual Salary: £65,000–£90,000, determined based on experience, technical expertise, and overall qualifications. Stock options may be offered where applicable.

Recruitment Process

CV Review → Initial Interview → Technical Test → Final Interview → Offer

About JIJ

JIJ is a global deep-tech company specialising in mathematical optimisation, artificial intelligence, high-performance computing, and quantum computing. Our mission: Make Society Computable,

Contributing to the Advancement of Humanity.

Our flagship platform, JijZept, provides an integrated environment for modelling, optimisation, benchmarking, and deployment. JIJ works with industrial customers, research institutions, universities, and technology partners across Japan, the UK, Europe, and North America.

We value people who are genuinely curious about science, algorithms, and complex systems; who balance scientific rigour with practical engineering; who take ownership of technical decisions; and who communicate openly across cultures and disciplines. You do not need prior quantum computing experience to join JIJ.

Equal Opportunity

JIJ is committed to creating an inclusive workplace where people with diverse backgrounds, experiences, and perspectives can thrive. We welcome applications from all qualified candidates regardless of race, ethnicity, nationality, gender, gender identity, sexual orientation, disability, age, religion, or any other protected characteristic.