Junior Medical Data Engineer

SOPHiA GENETICS; Chevanceaux; On-site; EUR 11,000 - 19,000; Full time

Benefits offered by this job

Monthly meal benefit

Paid internship

Job summary

SOPHiA GENETICS invites you to a six-month internship focused on applying Large Language Models to extract clinical information from medical reports. You will build a production-like workflow, compare models, and help shape future AI applications in healthcare.

You’ll generate synthetic reports, test prompts, and organize results into structured datasets while collaborating with clinical stakeholders and technical colleagues.

Qualifications

• You’re studying Data Science, AI, CS, or related field.
• You’re comfortable with Python for data processing, analysis, automation or ML.
• You understand SQL and relational databases.

Responsibilities

• Contribute to a real proof-of-concept project on LLM-based extraction.
• Create synthetic medical reports with known ground truth for testing.
• Experiment with prompts, models, and fine-tuning to compare performance.
• Organize extracted data into structured databases across reports.
• Document findings and share results with technical and clinical stakeholders.

Skills

Python

LLMs

Data processing

Data analysis

Machine learning

Education

Student in Data Science / AI / Computer Science

Tools

SQL

Job description

What if you could use AI to unlock valuable medical information hidden in thousands of documents and help turn it into insights that improve healthcare?

At SOPHiA GENETICS, we are building the future of AI-assisted, data-driven medicine. Our technology helps healthcare professionals make sense of complex data, with the ambition of improving outcomes for patients with cancer and rare diseases.

Every day, hospitals generate enormous amounts of valuable clinical information - diagnoses, treatments, biomarkers, imaging results and more. But much of this information remains locked inside PDF reports, making it difficult to analyze and use for research and innovation.

That's where you come in.

Join us for a six-month internship and explore how Large Language Models (LLMs) can transform unstructured medical reports into reliable, structured data that could support real-world evidence, clinical research and future AI applications.

What you'll work on:

You’ll contribute to a real proof-of-concept project, exploring how different LLMs can extract clinical information accurately, efficiently and at scale.

Here’s what that could look like:

• Create realistic synthetic medical reports. Generate synthetic PDF documents with known ground truth to safely develop and test AI models without using real patient data.
• Put LLMs to the test. Experiment with different models, prompting techniques and fine-tuning approaches to understand which perform best.
• Go beyond accuracy. Compare models based on reliability, robustness, processing speed and computational cost.
• Turn unstructured data into insights. Explore how extracted medical information can be organized into structured databases and connected across patient reports.
• Think about real-world applications. Depending on project progress, contribute… building a production-like workflow capable of processing incoming medical reports.
• Make your findings count. Document your work and share results and recommendations with technical colleagues and clinical stakeholders.

Requirements

You might be studying Data Science, Artificial Intelligence, Computer Science, Data Engineering or a related field.

You’ll probably enjoy this internship if:

• You’re confident with Python. You have hands-on experience using Python for data processing, analysis, automation or machine learning projects.
• You know how to work with databases. You have a good understanding of SQL and experience querying, managing or structuring data in relational databases.
• You have experience with AI and LLMs. You’re familiar with Large Language Models, their capabilities and limitations, and ideally have experimented with prompting, model evaluation or fine-tuning.
• You understand data engineering fundamentals. You’re comfortable working with different data formats, building data-processing workflows and transforming unstructured information into structured datasets.
• You’re curious about healthcare. You’re interested in how AI can support clinical research and medical decision-making. Knowledge of medical data or healthcare terminology is a strong advantage.
• You enjoy solving complex problems. You have strong analytical skills, attention to detail and a rigorous approach to testing and validating results.
• You can work independently. You’re comfortable taking ownership of a project, experimenting with different solutions and seeking feedback when needed.
• You communicate clearly. You can explain technical findings, model performance and trade-offs to colleagues from both technical and non‑technical backgrounds.

You don’t need to know everything about medical AI already. We’re looking for someone who is curious, rigorous and excited to explore how emerging technologies can solve real problems.

Benefits

What you’ll get as part of a team?

• Work on a meaningful AI project, with the opportunity to contribute to the development of future healthcare technologies.
• Gain hands‑on experience with LLM evaluation, fine‑tuning, synthetic data generation and AI workflow development.
• Learn from experienced colleagues working at the intersection of AI, data science and healthcare.
• Build your professional network in an international environment and participate in company events and activities.
• Enjoy a monthly meal benefit and a paid internship according to our internship practices and degree level.

Duration: 6 months

All SOPHiA GENETICS internships and apprenticeships are paid according to each country's best practices and degree level.

Bring your curiosity, challenge the models and help us explore how AI can turn complex medical data into meaningful insights.