Bioinformatics Software Engineer

About IMU Biosciences

IMU Biosciences has developed proprietary platform technologies that generate and translate vast system-level immune data into actionable insights and tools to drive the development of precision medicines across a variety of diseases. IMU leverages advanced immune profiling with proprietary AI and machine learning analytics to uncover novel clinical immune signatures. IMU continues to establish partnerships with leading pharma and biotech companies to advance disease diagnosis, optimise product selection, and improve patient stratification and monitoring - while also building its own pipeline of innovative products.

About The Role

We are seeking a Bioinformatics Software Engineer to drive the flow cytometry data engineering infrastructure, annotation pipelines, model registry operations, and code governance for an advanced clinical biomarker discovery platform operating under machine learning industry best practices and Medical Device Machine Learning (MedDev ML) standards.

In this role, you will be responsible for translating high-dimensional flow cytometry data into reproducible, production-grade features compliant with industry best practices and MedDev ML framework requirements. Your key deliverables will include frozen, reproducible feature pipelines and locked model registry artifacts. You will oversee automated cell-type annotation, manage model artifact registries, enforce data integrity, trace compliance, and handle leakage controls across machine learning partitions while leading peer code reviews.

Core Responsibilities

• Flow Cytometry Data Engineering & Annotation Pipelines: Execute automated cell annotation workflows on large-scale cytometry datasets; generate cell membership files, cell subset ratio tables, and automated outlier quality-control (QC) reports. Apply locked automated gating ensembles across cross-validation and evaluation partitions, tracking and recording per-patient gating exposure metrics. Establish and lock baseline models and numerical stability criteria for flow cytometry feature extraction pipelines.

• Model Registry & Governance: Manage the formal model lifecycle registry—versioning, freezing, and locking gating ensembles and candidate predictive models in adherence with industry best practices and MedDev ML regulatory frameworks. Refit shortlisted machine learning models on expanded cohorts and execute final model freezing prior to blinded clinical evaluations. Ensure full reproducibility by committing versioned code, pinning environment specifications (Conda/Docker), and logging all random seeds and model hyperparameters.