Назад
Company hidden
3 дня назад

Data Scientist (Responsible GenAI)

Формат работы
onsite
Тип работы
project
Грейд
senior
Английский
b2
Страна
Netherlands
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
Data Scientist (Responsible GenAI) (GenAI/Responsible AI): Building evaluation frameworks, metrics, and guardrails for GenAI assistants, report generation, document processing, and agentic applications with an accent on statistical evaluation, LLM-based assessment, and production-ready Python development. Focus on designing reliable experiments, measuring guardrail precision and recall, and integrating evaluations into CI/CD, APIs, and production workflows.

Location: Utrecht, Netherlands

No salary specified.

Company

hirify.global provides human-focused work across a broad portfolio of small and large clients in the banking and insurance sector.

What you will do

  • Build reusable evaluation frameworks and tools for repeatable testing, comparison, and regression analysis of GenAI solutions.
  • Develop and validate metrics using deterministic checks, model-based evaluation, and human assessments.
  • Design statistically sound experiments, datasets, evaluation scenarios, sampling strategies, and edge-case tests.
  • Develop and assess guardrails using rules, LLM-as-a-judge methods, and trained ML classifiers.
  • Measure guardrail effectiveness, including precision, recall, detection thresholds, and robustness.
  • Deliver tested, maintainable, production-ready code and integrate evaluations into CI/CD, APIs, and production workflows.

Requirements

  • Practical experience with LLM-based applications, RAG, prompting, grounding, and context management.
  • Strong Python skills and experience converting prototypes into maintainable, tested, production-ready code.
  • Experience with version control, automated testing, and CI/CD.
  • Applied statistical expertise in evaluation and experiment design, including sampling, dataset representativeness, and metric reliability.
  • Experience assessing and selecting rule-based, LLM-as-a-judge, and ML-classifier guardrails.
  • Knowledge of precision, recall, and class imbalance.
  • Ability to explain results to engineers and stakeholders and write clear technical documentation.

Nice to have

  • Experience with agentic evaluation, tool selection, and multi-step workflows.
  • Experience testing privacy controls, prompt-injection defenses, or other GenAI safeguards.
  • Experience with controlled adversarial testing or robustness assessment.
  • Experience with evaluation tooling, observability, or continuous evaluation in production.
  • Experience working in a regulated environment.

Culture & Benefits

  • Contract assignment from 01.01.2027 to 30.06.2027, with a possible extension.
  • 36 hours per week.
  • ZZP/freelance self-employment is not possible.
  • Work is part of the Responsible GenAI team in the banking and insurance sector.

Hiring process

  • hirify.global reviews the CV and discusses experience, availability, and rate.
  • Suitable candidates are proposed to the client, who usually responds within five working days about an interview.
  • The process may include a second or third interview, onboarding formalities, personal and passport details, Chamber of Commerce registration where necessary, and possible external screening.

Будьте осторожны

Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →