Axiombio Data Engineer, Life Sciences to build data pipelines and infrastructure for laboratory, chemistry, and clinical data supporting drug discovery and AI models. Develop quality controls and dashboards for data reliability and accessibility.
Responsibilities
Build production-grade Python and SQL pipelines to ingest, process, validate, and serve high-content imaging, omics, mass spectrometry, and ADME data alongside chemical and clinical datasets.
Own the shared data infrastructure that powers Axiom’s experiments, models, and scientific agents, including orchestration, storage, compute, isolated development environments, and continuous integration.
Unify compound identities, sample histories, experimental metadata, and assay results across laboratories and sources, with reliable schemas, traceable transformations, and reproducible dataset versions.
Develop statistically rigorous quality controls to detect missing data, anomalies, batch effects, and distribution shifts across imaging fields, wells, plates, and experimental batches.
Build dashboards, monitoring, and alerts that help scientists identify experimental failures, diagnose pipeline issues, and act on data quality problems quickly.
Give scientists, ML researchers, and AI agents reliable access to experimental data through well-designed tables, APIs, and analysis tools.
Partner with laboratory and computational teams to investigate data and model failures, improve experimental workflows, and deliver validated datasets for modeling, evaluation, and agent training.
What we look for:
Exceptional data engineering skills, strong computer science fundamentals, and a track record of building reliable, scalable systems with clear, maintainable code.
Deep statistical understanding. You can explain a method’s assumptions, assess whether they hold, and recognize when its results are misleading.
A strong instinct for correctness. You notice when numbers do not make sense, trace problems to their source, and build safeguards that prevent them from recurring.
The ability to find meaningful signals in messy scientific data, choosing methods that fit the experimental design and distinguishing biological variation from technical artifacts.
