skip to Main Content

NLP Architect @ Lynxcare



1.     Purpose of the position

The NLP team builds new NLP components with the newest technologies and support Operations doing customer projects. The NLP architect will redesign the process of building new NLP models, describe the tools needed for this, and will lead the deployment of the new process in R&D and Operations.
The Architect will initially focus on the improvement of the workflow of model training and deployment. He needs to improve the monitoring of the process, improve the generation of training sets and unseen sets, and make the process of training models more streamlined. Given the growth of the company, this streamlining is a high priority. 

2.    Reporting

The NLP architect reports to the NLP team lead. The NLP architect should be a good communicator, as the introduction of new workflows for building models requires a lot of coordination, both within the NLP team as with other R&D teams and Operations. In general all NLP team members work together in the team, but also have frequent collaboration with the data engineering team, the annotation team, and the business consultant.

3.    Tasks and responsibilities

•    The NLP architect initially will be mainly working on the project to propose and introduce a new training workflow for NLP models. It is a project that impacts multiple R&D teams and operations. Working together with multiple teams and having impact on them is a constant in the work of the NLP architect. So, a large part of the job is talking to people, sell proposals, and let the people behave as one large team when introducing the implementation of ideas. The architect should bring proven experience with training models. It will improve the efficiency of the team now that the number of customers is growing. 

•    The architect will also propose a new system for monitoring the model quality. He/she will have to convince the NLP team and the internal customers about the merits of this new way of working. In two years training workflow and quality monitoring model will have to be settled. The new hires will then work on other initiatives to support the expansion of LynxCare. This can be better support of other languages, … 

•    The architect and the NLP team should continuously improve the Machine Learning processes for a better monitoring of the model development.

•    The architect and the NLP team train models for new use cases in several languages and deploy models to Azure cloud.

•    The architect has a rather complex collaboration with other stakeholders. The NLP architect has many contacts with the NLP team, with other R&D teams (data engineering, cloud infrastructure, and application development) and with the different roles in Operations. 

•    The architect stays up to date with latest innovations in NLP

4.    Reason for the job opening

LynxCare’s business is growing fast. To better support this growth the NLP team is extended with an NLP Architect.

5.    Expectations first 12 months

It is the expectation that the training workflow is replaced in lockstep with the change of NLP components. This means that a whole new workflow and accompanying tools are in use within 12 months. 



  • MSc or PhD degree in a quantitative field (e.g., artificial intelligence, computer science, biostatistics, biomedical informatics,       applied mathematics, or similar).
  • At least 2 years experience with training and deploying ML/NLP models, preferably in the healthcare sector.
  • Solid and proven working experience with machine learning, deep learning, and natural language processing.
  • Good experience with organizing workflows. Knowledge of MLFlow etc is required.
  • Being solution oriented and thinking in options rather than limits and barriers, able to enthuse colleagues on the power and benefits of leveraging technologies in their business. Fostering personal relationship without wasting their time. 
  • Dedicated/committed to quality as well as timely execution. High sense of responsibility.
    Being open, cards on the table, honest, respectful in communication and not political. 
  • Enjoyable personality to collaborate with.


✔    Attractiveness of the sector (Healthcare, societal impact).
✔    Considered as one of the companies in the market that is shaping the sector/industry of Real World Evidence.
✔    Small, fast growing company, not a large corporate filled with internal politics. You will be part of a young team of 30+ colleagues, consisting of a mix of clinicians and data scientists, and you will work on challenging projects together with physicians and pharmaceutical companies.
✔    You will have the opportunity to help shape the company’s future.
✔    Most advanced in terms of technical capabilities (this is what customers are telling). A strong technical and operational team with extensive knowledge of the technology and the environment in the industry.
✔    It is a major opportunity for an NLP Architect to be working with the latest technologies, such as multilingual transformer models. The NLP architect will recreate the NLP workflow for building new models, which has an important impact on the success of the company in the next years. It is a cutting edge NLP function.
✔    The NLP architect can grow to NLP team lead or project manager as well.
✔    Hybrid working; partly in the office and partly work-from-home.
✔    A complete and competitive salary package with company car and fringe benefits
✔    Trainings & seminars with your team.


NLP Architect @ Lynxcare

LynxCare, with offices in Belgium (Leuven) and the United States (New York), is a Belgian Healthtech  scale-up and big data specialist in healthcare, which developed a AI- and NLP  powered SaaS clinical (all-in-one) data platform for hospitals and for life sciences research.

Lynxcare is currently available in Belgium, The Netherlands and the US. The company is continuously developing new use cases covering more disease areas (with spearheads in Cardiology and Oncology) and additional languages (currently active in English, Dutch, French, and German) thereby widening its market reach, aiming at expanding into the French & German market as of 2022.

Examples of customers are: UZA, UZ Leuven, ASZ Aalst, AZ Maria Middelares, AZ Geel, AZ Groeninge, Astra Zeneca, Takeda, Pfizer, ARJO, Janssen, Bayer,… 

LynxCare's platform assists hospitals with time-consuming and complex patient data processing and hospital operations. The AI platform offers benefits to patients, physicians and healthcare staff alike. Silos are broken, data is made more insightful and the platform provides more structure. The platform allows insights to be drawn across the multitude of databases and data structures in the hospital. From structured laboratory studies to unstructured reports (with NLP). Patients also get much better insights into the success rate of certain procedures. 

Whatsapp op jouw computer?

Amandine ParisisBereikbaar van maandag tot vrijdag tijdens kantooruren
Amandine ParisisHallo 👋
Meer weten over deze vacature?

Geen Whatsapp? Stuur ons een berichtje


  • Sleep bestanden hierheen of
    Toegestane bestandstypen: jpg, gif, png, pdf, doc, docx, txt, Max. bestandsgrootte: 20 MB, Max. aantal bestanden: 5.

    Back To Top