ML Engineer

Il y a 2 jours

Brussel Hoofdstad, Belgique Excelra Knowledge Solutions Pvt. Ltd. Temps plein 90 000 € - 120 000 €/an

We are hiring a Scientific Manager to take ownership of the machine learning and experimental design pipelines behind our customer’s genomic target prediction work. Much of this currently lives in individual scripts maintained by individual scientists. Your job is to turn it into an automated, version-controlled and reproducible platform the whole team can rely on – and to keep it scientifically sound by working shoulder to shoulder with our customer’s wet-lab and computational biology colleagues.

This is a hands‑on engineering role in a scientific environment. You will spend more of your time building, automating and hardening pipelines than running individual analyses, so we are looking for someone who is as comfortable with Git, containers and workflow managers as with genomics data.

A PhD in bioscience engineering, bioinformatics, computer science or machine learning engineering – or an MSc with 5+ years of equivalent experience. You can demonstrate this through peer‑reviewed publications, contributions to public code repositories or relevant project work.

  • Strong Python, and the habit of writing code that colleagues can read, test and re‑run.
  • Proven experience taking analysis code into production: version control with Git, code review, packaging, containerization (Docker, Singularity/Apptainer) and automated testing or CI/CD.
  • Hands‑on experience with workflow orchestration – Nextflow, Snakemake or equivalent.
  • Experience running workloads on cloud infrastructure (AWS) and/or HPC clusters (e.g. SLURM).
  • Solid grounding in statistical modelling and machine learning applied to biological data, including model evaluation and experiment tracking.
  • The ability to translate between engineering and bench science, and to work with scientists who are not engineers.
  • You are eligible to live and work in Belgium.

What will you do?

  • Automate and professionalize internal ML and experimental design pipelines for genomic target prediction.
  • Migrate existing scripts and notebooks into a production‑ready, version‑controlled codebase.
  • Reduce the manual input required for dataset preparation, settings configuration and pipeline execution.
  • Integrate pipelines into existing digital infrastructure, including AWS and HPC environments.
  • Implement version control, testing and reproducibility standards across ML workflows, and help the team adopt them.
  • Collaborate closely with wet‑lab and computational biology colleagues to make sure pipelines meet real scientific requirements.
  • Run several projects in parallel, (re)assigning tasks as priorities shift, and present results concisely to stakeholders inside and outside your domain.