A
NLP Engineer – Document Processing
Amara Diallo
Full-time · Mid-level · Amsterdam
About the role
We will not pretend this is an easy role. We process legal and financial documents in 9 languages. Our current extraction pipeline uses a mix of regex, older Hugging Face models, and a lot of post-processing code that has grown organically over two years. It works, but not well enough for the enterprise clients we are trying to close. We need an NLP engineer who has dealt with messy, real-world document pipelines and knows how to improve them systematically — not someone who will arrive with one new model and assume it fixes everything.
Responsibilities
- Audit the existing extraction pipeline and document current failure modes
- Fine-tune or replace extraction models where gaps are clearly identified
- Build automated evaluation datasets across document types and languages
- Improve post-processing logic to handle edge cases more robustly
- Work with the product team to define acceptable accuracy thresholds per use case
Requirements
- 3+ years NLP engineering with document processing systems in production
- Strong Python and Hugging Face Transformers
- Experience with named entity recognition, document classification, or information extraction
- Able to evaluate model performance per entity type and per language
- Comfort working with legacy codebases — improves what exists, doesn\'t throw everything away
Benefits
- Technically challenging, commercially relevant NLP work
- Full remote
- €75,000 – €95,000
- Conference budget
- Small team, direct impact
Job Type
Full-time
Level
Mid-level
Language
English
Salary Range
€75,000 – €95,000
AI Expertise
NLP & Prompt Engineering
Ready to apply for this role?
Create a free talent account in under 2 minutes.
- Apply to verified AI companies
- Get AI-matched job recommendations
- Message hiring managers directly
- Build your public AI talent profile