EmergingLanguage and multilingual intelligence

Low-Resource Language Data Researcher

Also called Low-Resource NLP Scientist

A Low-Resource Language Data Researcher builds ethically sourced datasets and evaluations for languages that models serve poorly. It matters because it widens who a product works for, and shows where the model really fails.

What they do

The work starts with a language and a community, not a dataset. They map what text and speech exist, who owns it, and who should be asked before it is used. They design the corpus: sources, dialects, scripts, spelling variation and domains. They agree consent terms with community partners and set up review by native speakers. They check how tokenisers split the language and where that hurts. They train or fine-tune small models to test whether the data helps. They then build an evaluation set that reflects real use, and write up its limits.

What a strong hire delivers

In the first months, a strong hire delivers a clear plan for one or two target languages: who the partners are, what consent covers, and what the corpus will contain. They produce a first community-reviewed sample and a baseline evaluation that shows where current models fall short. They also leave behind documentation that lets others reuse the data safely and understand what it can and cannot support.

How you can tell they are good

Ask them to walk through one dataset they built from first contact with a community to release. Listen for who reviewed it, what they changed, and what was left out. Ask how they would judge utility beyond a headline score. A strong candidate talks about dialect, orthography and consent in specifics, and admits where their benchmark is weak.

Why it is hard to fill

The title barely exists. The work usually sits inside NLP research, linguistics projects, translation teams or nonprofit language technology, under other names. Few people combine model skills with real community relationships, and relationships cannot be hired in. Candidates are often strong on one side only: either technical or fieldwork. Language coverage matters too, so the right person for one language may not suit another.

Where to find them

Look in NLP research groups working on multilingual or under-served languages, university linguistics and language documentation projects, and nonprofit or community-led language technology initiatives. Translation and localisation teams hold useful people, especially those who have run native-speaker review. Some come from speech teams, digital humanities, or a model provider's multilingual evaluation group. Many hold titles such as research engineer, computational linguist or data scientist.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Low-Resource Language Data Researcher

  1. Describe a dataset you built for an under-served language. How did you source it and who reviewed it?

    A strong answer names sources, consent arrangements and the review process, and says what reviewers changed. Worry if the answer is only scraping and filtering, or if community involvement is vague or came after the fact.

  2. How would you decide whether a community consents to their data being used in a model?

    Good answers cover who speaks for a community, informed and revocable consent, and clear terms on reuse. They accept that consent can be slow. Worry about a single signed form treated as enough.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building language technology that works for people whose languages are often ignored. We are hiring a researcher to create ethically sourced data and evaluation for under-served languages, working directly with the communities who speak them. The outcome is data and benchmarks those communities have reviewed and trust.

What you will do

  • Design corpora for target languages, covering dialects, scripts and domains
  • Build and maintain partnerships with community organisations and speakers
  • Agree consent terms and document how data may be used
  • Run native-speaker review and act on what it finds
  • Build evaluation sets that measure real language utility
  • Test tokenisers and fine-tuned models against those sets
  • Publish clear documentation of coverage, limits and provenance

What you will bring

  • Shown work building a dataset or benchmark for a low-resource language
  • Experience working with speaker communities or native-speaker reviewers
  • Practical skill with transformer models and tokenisers
  • Sound approach to evaluation, including its limits
  • Grounding in linguistics, translation or language technology
  • Clear written communication of data decisions

Nice to have

  • Fluency in a under-served language
  • Experience with speech or multilingual evaluation
  • Prior work on data governance or community licensing

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Transformer modellingAdvancedImplement attention-based sequence or multimodal models
  • Representation learningAdvancedLearn transferable features or embeddings
  • Linguistics and localisationWorkingAdapt language behaviour across cultural and linguistic contexts
  • Model evaluationWorkingMeasure capability and failure modes credibly

Other roles in language and multilingual intelligence

Hiring nowNatural Language Processing ScientistAn NLP Scientist builds and tests methods for language understanding and generation. They make sure a company's language models work across…Hiring nowMultilingual Model ResearcherA Multilingual Model Researcher improves how well a model transfers and behaves across languages. It matters because a product that works in one…Hiring nowLanguage Localisation EvaluatorA Language Localisation Evaluator judges whether model output means the right thing, sounds right and is appropriate in a specific locale. It…

Related roles in other families

EmergingReasoning Research ScientistReasoning, planning and searchA Reasoning Research Scientist tests and builds methods that make a model's multi-step inference reliable. They show the company which approaches…Hiring nowPlanning Algorithms ResearcherReasoning, planning and searchA Planning Algorithms Researcher builds planners that choose sequences of actions under hard constraints. Companies hire them so agents and systems…EmergingInference-Time Compute ResearcherReasoning, planning and searchAn Inference-Time Compute Researcher decides how much thinking a model should spend on each problem. They raise accuracy on hard tasks while keeping…EmergingNeuro-Symbolic AI ResearcherReasoning, planning and searchA Neuro-Symbolic AI Researcher combines learned models with symbolic rules and solvers, so systems reason consistently where rules matter. That…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Language and multilingual intelligenceAll new job families

Start a search

Hiring a Low-Resource Language Data Researcher?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.