pro
Klaas Lieuwes
Dutch language, data & ERP support for teams
- $1k+
- Earned
- 1x
- Hired
- 1
- Followers
Validating and correcting PII annotations in synthetically generated Dutch-language conversations, covering both nl-NL and nl-BE (Flemish) locales.
The work goes beyond simple labelling. Each conversation is reviewed for missing or incorrect entity spans, offset accuracy, and whether every instance of a PII entity is consistently tagged across turns. On top of that, the Dutch language quality is assessed — grammar, naturalness, locale-appropriateness, and conversation coherence — with corrections and explanations provided where needed.
PII entity types covered include names, addresses, phone numbers, email addresses, IBANs, ID numbers, dates, usernames, and passwords, each validated against the correct Dutch or Belgian format.
Quality targets: inter-annotator agreement above 95%, full locale validity, and self-QC accuracy of 98% or higher.
Native Dutch speaker covering both standard Dutch and Flemish — not two separate skill sets, but one annotator who knows the difference.
1
19
Transcribing and annotating Dutch-language audio for AI speech recognition training at Deepgram, one of the leading speech AI companies. Work involves listening carefully, producing accurate transcripts, and applying consistent annotation according to strict quality guidelines.
This is specialist work — not bulk transcription, but quality-controlled annotation used directly to train production speech models. Accuracy and consistency matter more than speed.
Native Dutch speaker with genuine familiarity with regional speech patterns, accents, and edge cases that a non-native annotator would likely miss or flag incorrectly.
1
23
Dutch Website Translation & Technical Content
Translating website copy, product descriptions, and technical documentation from English into Dutch for an international client operating in the technology sector.
The work requires more than language skills alone — understanding the underlying product, its technical context, and the target audience is essential to producing translations that are accurate, natural, and usable. Generic machine output doesn't cut it here; the client needs a native speaker who can think along.
What this project covers:
Website UI copy and product pages
Technical documentation and support content
Ongoing collaboration on an hourly basis — not a one-off, but a sustained working relationship
Good fit for clients who need:
Native Dutch, not translated Dutch
Someone who understands what they're translating, not just the words
Reliable turnaround and consistent terminology across deliverables
Hourly — ongoing availability.
1
31
ERP parts data structuring for an electric vehicle manufacturer
1
1