TL;DR
Principal Applied Scientist (AI): Leading the development of next-generation grounding services powering AI applications with an accent on retrieval, attribution, and reasoning methods. Focus on designing scalable production models, building citation systems, and collaborating cross-functionally to deliver trustworthy, explainable AI solutions.
Location: Beijing, China, onsite with expected office presence
Company
hirify.global builds advanced AI platforms and services to empower organizations worldwide with innovative technology and inclusive culture.
What you will do
- Own the science roadmap for grounding, including retrieval, re-ranking, attribution, and reasoning methods.
- Build citation and provenance systems to reduce hallucinations and increase user trust.
- Advance tool-augmented grounding with schema-aware retrieval and real-time data connectors.
- Collaborate with product, engineering, and research teams to productionize scalable AI models and APIs.
- Mentor applied scientists and data scientists, establishing best practices in experimentation and error analysis.
- Communicate progress and contribute to ethics, privacy, and bias mitigation policies.
Requirements
- Location: Must be based in Beijing, China with onsite work expected
- Bachelor’s degree or higher in relevant fields with 4+ years experience in search, retrieval, or ranking systems.
- Proven experience shipping LLM-powered or Retrieval-Augmented Generation systems into production.
- Strong coding skills and machine learning foundation.
- Ability to lead through ambiguity and deliver measurable impact in fast-paced environments.
- English: Proficient (B2) required
Nice to have
- Advanced degrees with extensive experience in statistics, econometrics, or computer science.
- Publications in top-tier AI and information retrieval conferences.
- Hands-on experience with LLM development, pretraining, fine-tuning, and reinforcement learning.
- Contributions to open-source LLM inference frameworks.
Будьте осторожны: если вас просят войти в iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →