Providing daily over a million real-time predictions to our user base, the AI team has been instrumental in launching our flagship product, Lifen Sending, which facilitates the seamless exchange of medical notes among healthcare practitioners in 800 healthcare organisations in France. Over the past 6 years dealing with Medical Data, our team has built a strong expertise in building robust models that deliver actionable value to our users by accelerating laborious human-intensive administrative workloads with high standards in terms of security and privacy.

Our current team (3 ML Engineers & 1 Datascience Manager) has built Layout Extraction Models for extracting text from documents and preprocessing them for downstream tasks, Visual Explainable NLP Tools for the support team & users, Pseudonymisation algorithms to be able to work on algorithms without exposing private medical data and finally our core Information Extraction Models from documents (Classification, NER, etc.).

Lifen is now investing in a new offer for Healthcare Organisations and academic & industry Research Partners that will tap into our expertise in dealing with medical documents and Lifen’s Reach (800 Healthcare Organisation & 3M documents sent per month) to structure real-world research-grade data from the medical reports handled by Lifen: Lifen Cohort. We already have a proof of concept working with generative AI on a first production project. It enhances this even more human-intensive workload of structuring data from health records to change the scale and quality at which medical research can be done!

We are seeking a forward-thinking, solution-oriented, autonomous Machine Learning Engineer to expand our team of 3 ML Engineers to help with historic algorithms and develop new ones into our new products

At Lifen we have a ‘broad’ definition of Machine Learning Engineer, which encompasses the complete lifecycle of algorithms : from proof of concept to integration into our technology ecosystem, production, and maintenance, while adhering to software engineering best practices.

You will be actively involved in production-level code development, instead of conducting isolated experiments with Jupyter notebooks.

What you can expect :

- Adapt existing algorithms to novel problems and develop unique solutions as required ;

- Ensure the performance of our core models by taking part in their regular maintenance to meet product needs ;

- Contribute to our ML training infrastructure by developing new features, thereby aiding model production and evaluation ;

- Actively follow, experiment and integrate what’s currently happening with generative AI to shape our roadmap with our lead ML engineer, from prompt engineering to LLM finetuning.

⛓️ In terms of development, you have a software engineering mindset and proficiency in modern Python and the ecosystem (Poetry, MyPy, FastAPI, Pydantic, etc.) ;

🔮 Regarding machine Learning, you possess +3 years experience with Tensorflow, Keras, and Huggingface (and played with OpenAI);

🎶 When it comes to orchestration, you've developed knowledge of MLOps tools like Kubeflow ;

⚙️ Operations: You are comfortable with production & continuous integration concepts (Git, Docker, CircleCI, Monitoring, etc.) ;

🪐 You already have a proven experience in deploying machine learning pipelines in production settings, from data collection to model evaluation ;

📇 In general, you have a solid understanding of the Natural Language Processing field ;

👥 Excellent communication skills, autonomy, adaptability, attention to detail, and team spirit ;

🇫🇷 We expect from you to be completely fluent in French as it is our main spoken language ;

🎂 If you have knowledge of the healthcare industry, well that's the icing on the cake.

- HR Interview with Héloïse, HR Business Partner ;

- Interview with Lead Machine Learning Engineer ;

- Technical Assessment ;

- Final Round with Félix, Head of Data Platform ;

- Reference Check.