REMOTE Data Scientist

We are currently looking for a Data Scientist to help us with automation information extraction from prescription documents. The processing is based on OCR and an ML model for recognizing relevant fields and values. A current focus is on improving data and evaluation quality. The biggest challenges are inconsistent and sometimes erroneous training and reference data, different notations and formats of the extracted values, and a partially documented existing solution. Field values ​​must be standardized to ensure reliable comparison between model predictions and reference data. Data and labeling issues will be addressed and resolved in collaboration with the relevant department. Additionally, LLM-generated weak labels will be reviewed in coordination with a task force and used to further expand the data foundation.


Project details:
·      Start date: ASAP
·      Duration: 31.12.26
·      Full-time: 40 h/week
·      Location: Remote/Munich
·      Language: German (MUST)

Experience:
•        Excellent Python and Pandas skills
•        ML, ideally in OCR, Document AI, or information extraction
•        Data cleansing and validation
•        Understanding of evaluating ML models and building robust quality metrics
•        Basic understanding of Databricks and data-driven development processes
•        Experience with LLMs, synthetic data, or weak labeling is advantageous
 
Would you be interested in learning more about the project, we will be happy to discuss it with you!