F

Data/ML engineer, fusion data

Fusionality·Lausanne, ch·22.10.2026

 join.com

 
 
Festanstellung80–100%
Required language
English conversational, French conversational
Job written in
English
Location
Lausanne, ch
Work type
On-site

 

Fusionality is a spin‑out from EPFL's Swiss Plasma Center that develops the measurement, simulation and control stack for future fusion energy reactors. Backed by leading investors and founded by experts in fusion physics, advanced control and AI, the company is assembling a founding engineering team in Lausanne. The role is a Data / Machine Learning Engineer focused on turning large, heterogeneous experimental data into clean, labelled, trustworthy datasets for plasma monitoring and control. Day‑to‑day you will design, build and operate pipelines that ingest experimental fusion data and produce high‑quality, versioned datasets. You will create ML‑assisted labelling and event‑detection tools, using weak supervision, active learning and anomaly detection, to identify disruptions, MHD activity and diagnostic faults. Ownership of data quality, provenance and calibration metadata will be central, as will providing easy access to the rest of the team. Collaboration with physicists, computational scientists, control engineers and external customers will be required to validate datasets and integrate their data streams. The position demands hands‑on experience constructing data pipelines for large scientific or industrial time‑series formats such as HDF5, Parquet or Zarr. You must be proficient in designing, querying and optimizing both SQL and NoSQL databases, with familiarity in time‑series stores like TimescaleDB or InfluxDB. Applying machine‑learning techniques to time‑series data in Python, using PyTorch or JAX for sequence models, anomaly detection or semi‑/weakly‑supervised labelling, is essential. Strong software‑engineering habits, including testing, reproducibility and data versioning, are expected. The employer looks for evidence of independent, high‑level work, such as a PhD thesis, published code adopted by others, or an owned industry project, and the ability to design data architectures from day one. Proficiency with agentic coding and modern AI tools to accelerate development is also required. Nice‑to‑have skills include experience with human‑in‑the‑loop labelling tools, pipeline orchestration frameworks (Dagster, Airflow, Prefect) and data‑versioning solutions (DVC, lakeFS). Backgrounds in large‑scale physics experiments, fusion, accelerators, astronomy, or familiarity with fusion data systems such as MDSplus or IMAS are advantageous. The role offers a key position in a fast‑growing startup, the chance to shape cutting‑edge technology for real fusion devices, and a work environment in Lausanne, Switzerland, known for high quality of life, competitive salary, equity and immigration assistance. What the role asks for: - Build pipelines for large scientific time‑series (HDF5, Parquet, Zarr) - Design, query and optimise SQL and NoSQL databases - Experience with TimescaleDB or InfluxDB - Apply ML to time‑series in Python (PyTorch or JAX) - Strong software‑engineering habits: testing, reproducibility, versioning - Independent work evidence: PhD thesis, published code or owned project - Design data architectures independently from day one - Proficient with agentic coding and modern AI tools - Human‑in‑the‑loop labelling tools (nice-to-have) - Pipeline orchestration (Dagster, Airflow, Prefect) (nice-to-have) - Data versioning tools (DVC, lakeFS) (nice-to-have) - Experience with fusion or large physics experiment data (nice-to-have)

 

 

 

 

 

Kostenlos

Engineer: Lebenslauf-Vorlage

Bewirb dich mit einem Lebenslauf im Schweizer Aufbau — mit Beispieltext und den Anforderungen, die Engineer-Inserate am häufigsten nennen.

Lebenslauf-Vorlage Schweiz ansehen

Engineer: was der Markt gerade verlangt

Engineer-Stellen gehören zu den regelmässig ausgeschriebenen Berufen auf SwissJobs.app.

Weitersuchen

Ähnliche Jobs per E-Mail erhalten