Назад
Company hidden
6 часов назад

Senior AI Engineer (Quality & Knowledge)

Формат работы
remote (только Europe)/hybrid
Тип работы
fulltime
Грейд
senior
Английский
b2
Страна
France/Spain
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
Senior AI Engineer (Quality & Knowledge) (LLM agents/payroll AI): Building and evaluating user-facing AI agents that help companies prepare, check, and run payroll with an accent on knowledge systems, retrieval quality, and workflow reliability. Focus on investigating failures across models, context, knowledge, and tools, improving agent harnesses and guardrails, and measuring regressions in production.

Location: Hybrid in Paris, with remote work available from anywhere in France or Spain.

Company

hirify.global provides a SaaS payroll and HRIS solution for SMBs, operating across France, Spain, and the UK.

What you will do

  • Work with Customer Support and payroll experts to convert complex real-world cases into testable AI behavior.
  • Build evaluations using representative cases, expert reviews, and production feedback to assess answers, actions, and workflow outcomes.
  • Investigate failures across models, context, knowledge, retrieval, and tools, then implement and measure improvements.
  • Contribute to specialized agents, orchestration, and harness engineering, including context management, reusable tools, guardrails, and human handoffs.
  • Improve knowledge pipelines and retrieval quality, using regression evaluations and observability to maintain reliable workflows.

Requirements

  • 7+ years of software engineering experience, including designing and operating backend services or pipelines in production.
  • Hands-on experience building LLM applications or agents, including tool integration, context management, and handling uncertainty.
  • Practical experience evaluating LLM systems with representative datasets, clear criteria, and automated or human review.
  • Strong backend engineering skills and the ability to contribute to a TypeScript/Node.js codebase.
  • Ability to work directly with domain experts and translate ambiguous cases into testable behavior.
  • Care for security, permissions, observability, and maintainability.

Nice to have

  • Experience with Langfuse or similar tools for tracing, prompt management, datasets, and experiments.
  • Familiarity with AI SDK or comparable frameworks for tool-using agents.
  • Experience with hybrid search, reranking, OpenSearch, or other search and vector technologies.
  • Experience measuring judge agreement, evaluation uncertainty, or agent consistency across repeated runs.
  • Experience building annotation tools or review workflows for domain experts.
  • Familiarity with AWS, data pipelines, and CI/CD for AI systems.
  • Experience in payroll, HR, customer support, or another domain where factual accuracy and traceability matter.

Culture & Benefits

  • Hybrid work in Paris with remote work possible from anywhere in France or Spain; additional requirements may apply.
  • Learning platform, professional development opportunities, and internal mobility.
  • Health insurance with 60% covered by hirify.global, transportation or sustainable mobility support, and a €9-per-workday restaurant card covered at 60%.
  • Work Council allowance, vacation bonus, MacBook, and parental support program.
  • 25 days of holidays plus RTT days depending on the contract.
  • Inclusive workplace with reasonable adjustments available for disabled candidates and employees.

Hiring process

  • Five stages: Talent Acquisition call, Hiring Engineering Manager meeting, technical interview with squad engineers, Product Manager and AE Designer/CS expert meeting, and final interview with the Tribe Engineering Director.
  • After the offer, an ISO27001 background check verifies identity, diploma, and past professional experience.

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 →