Senior NLP Data Science

Barcelona - Data Science - Full time

Last update: July 15, 2020

About the job

The work consists essentially of designing, implementing, and training algorithms that extract information from the clinical text, as well as collaborating in the development and maintenance of the main products of our company. We don’t draw a hard line between our research and engineering teams: they both research and develop market-quality software, putting previous research into practice. We mainly work in Python, using our own and existing NLP libraries.

About you

You must have skills in text mining or other non-trivial NLP tasks, and feel comfortable programming seriously (you must have Git skills, experience with libraries like pandas, scikit, tensorflow, and know how to create reusable and extensible code).

We are looking for someone who likes teamwork, teaching and learning from others, and talking about what you’re working on, with both technical and non-technical team members. In addition, you will be able to prepare scientific communications about the work he has done.


It is important that you feel comfortable helping the rest of the team to get the job done. However, the position is about:

  • Creation and testing of NLP models oriented to medical text processing. The tasks are, among others:

    • Named Entity Recognition
    • Named Entity Linking
    • Word-sense disambiguation
    • Language Modeling
    • Text Classification
    • Relation Extraction
  • Adaptation of the models to new languages.

  • Collaborate in writing papers and posters.

  • Adaptation of the models to production environments.

What we are looking for

  • Knowledge of Python and useful libraries for data science (pandas, numpy, sklearn, etc.)
  • General knowledge of Machine Learning.
  • Experience implementing neural networks (NN) and creating new NN models, hyperparameter adjustment, etc. Experience with a NN library (PyTorch / Keras / Tensorflow/ theano/ others).
  • Experience implementing deep learning models.
  • At least three years of professional experience working as a data scientist or data engineering.
  • At least two years of professional experience in Natural Language Processing.

What we offer

  • Negotiable salary
  • Full-time permanent contract.
  • Company profit-sharing scheme.
  • Flexible schedule, with the possibility of home office once a week.
  • A warm, transparent, and supportive team, with a huge emphasis on work-life balance.
  • Most days, lunch together in our sunny terrace.
  • The opportunity to make your mark in e-health and AI.


IOMED is a technological company of software development. It was launched in 2016, funded by local and international ventures.

We are passionate and talented young professionals, from all around Spain and the world (It couldn’t be any other way, as we’re based in beautiful and bright Barcelona). Our “dream team” is made up of mathematicians, statisticians, bioinformaticians, and physicians.

We are looking for people who are eager to innovate and be part of a project with an impact on the healthcare industry, enjoying what we do, team-work, and taking on new challenges.

IOMED is an equal opportunity employer. We are still a small team and are committed to growing in an inclusive manner. We want to augment our team with talented, dynamic people irrespective of race, color, religion, national origin, sex, physical or mental disability, or age.

What we do

Nowadays, around 50% of Clinical Trials are delayed due to patient recruitment, since patient data collection is performed in a manual fashion. As a result, clinical research is highly inefficient both in time and cost, taking years and billions of dollars to develop a new drug.

This problem could be solved through Real World Data, i.e. derived from Electronic health records (EHR). But unfortunately, up to 85% of existing clinical data is unstructured, i.e. in plain text. This also leads, in part, to the existence of data silos, making it impossible to aggregate data from different hospitals.

IOMED has found the solution to this situation, making it possible to take advantage of the full value of clinical Real World Data. We developed a tool that extracts the necessary data from clinical texts, which results in a structured, standardized, and interoperable database that contains the complete clinical information from hospitals.