TL;DR
Applied Scientist (AI): Designing and implementing new relevance metrics, building labeling pipelines, and fine-tuning language models to improve local search quality. Focus on prompt engineering, applying modern LLM techniques like Retrieval Augmented Generation, and creating scalable workflows for measurement and evaluation.
Location: Barcelona, Spain. Expected to work from the office at least four days per week if living within a 25-mile commute of a designated Microsoft office.
Company
hirify.global aims to empower every person and every organization on the planet to achieve more, fostering a culture of inclusion, respect, integrity, and accountability.
What you will do
- Design and implement new relevance metrics to measure and improve local search quality.
- Develop and optimize LLM/SLM labeling pipelines for consistent quality judgments.
- Engineer and fine-tune prompts for LLMs to enhance query understanding and classification accuracy.
- Apply modern LLM techniques such as retrieval-augmented generation for improved grounding and relevance.
- Build scalable workflows and dashboards for measurement, evaluation cycles, and quality checks.
- Analyze failure modes and improve prompt rubrics to reduce defect rates.
- Collaborate with cross-functional teams to integrate metrics and labeling systems into production environments.
Requirements
- Bachelor’s, Master’s, or Doctorate Degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or a related field, along with hands-on experience.
- Ability to meet Microsoft Cloud Background Check security screening requirements upon hire and every two years thereafter.
Culture & Benefits
- Accelerate career growth by working on state-of-the-art AI systems.
- Develop deep expertise in prompt engineering and model tuning.
- Hone skills in building robust data pipelines and quality frameworks.
- Work in a culture of inclusion, respect, integrity, and accountability.
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 →