Data Scientist, Lab & Protein Data

Adaptyv
Switzerland
Workplace: OnsiteFull timeFunction: Data Science & Machine LearningSkills: ["Interdisciplinary collaboration","Statistical rigor","Problem-solving","Judgment","Build mindset"]

Build the data science layer of an automated biotech lab “foundry,” turning raw, messy assay readouts into clean, trustworthy structured data. Define “good data” per assay type, automate quality checks, and develop QC/anomaly detection models to catch drift and artifacts. Partner with software and ML teams to improve data pipelines, connect results to protein sequence/structure, and deliver benchmark-grade datasets for customers and AI protein-design models.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Adaptyv
Adaptyv
2 months ago

Data Scientist, Lab & Protein Data

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 19 hours agoStatus: Live

Job Summary

Build the data science layer of an automated biotech lab “foundry,” turning raw, messy assay readouts into clean, trustworthy structured data. Define “good data” per assay type, automate quality checks, and develop QC/anomaly detection models to catch drift and artifacts. Partner with software and ML teams to improve data pipelines, connect results to protein sequence/structure, and deliver benchmark-grade datasets for customers and AI protein-design models.
Location: Switzerland
Workplace: Onsite
Employment Type: Full time
Job Function: Data Science & Machine Learning

Key Responsibilities

  • •Own the scientific logic of data quality across the foundry by defining what “good data” means per assay type and translating it into automated checks.
  • •Build anomaly detection and QC models to catch bad data (e.g., drift, variability, plate effects) and distinguish real signal from noise.
  • •Specify, review, and improve automated data pipelines with software/ML teams, feeding precise requirements for flags, auto-rejects, or human review.
  • •Connect experimental results to protein data (sequence, structure, design) so wet-lab and computational models reinforce each other.
  • •Turn foundry output into structured, documented, benchmark-grade datasets and ensure results are interpretable and comparable across runs.

Key Requirements

  • •Strong data science and bioinformatics background, fluent in Python (pandas, numpy) and comfortable owning experimental data end to end.
  • •Biology grounding to understand proteins, assays, and sequence/structure/function and interpret what the data means.
  • •Statistical maturity in process control, anomaly detection, and handling variability and batch effects.
  • •Proven builder who has shipped pipelines, tools, models, and datasets end to end in real use (not only prototypes).
  • •Interdisciplinary, AI-native approach; comfortable building across lab, software, and ML and using coding agents (e.g., Claude Code) where appropriate.
Experience:BiotechnologyBioinformaticsProtein/sequence-structure dataExperimental dataMachine learningAI protein design
Skills:Interdisciplinary collaborationStatistical rigorProblem-solvingJudgmentBuild mindset
Tech Stack:PythonPandasNumpyClaude Code

Company Brief

Adaptyv
Biotechnology company developing adaptive immunotherapy platforms and precision biologic treatments aimed at enhancing immune responses against disease. Focuses on engineering biologic solutions and translational research to accelerate therapeutic candidates from discovery to clinical development.
Industry: Biotech
Website