Qa Engineer, Localization

Zendesk Zendesk · Enterprise · Lisbon, Portugal +1

QA Engineer focused on localization, with a strong emphasis on testing AI-driven features and LLM-generated content. The role involves implementing AI quality frameworks, validating linguistic and functional correctness across locales, optimizing vendor operations, and collaborating on internal testing tools. Requires hands-on experience testing AI features and foundational scripting knowledge.

What you'd actually do

  1. Actively implement and run next-generation AI quality frameworks and evaluation tools to automate and scale our localisation testing capabilities.
  2. Design and execute test plans specifically for AI-driven features, validating LLM-generated content for linguistic accuracy, cultural appropriateness, and functional correctness across all supported locales.
  3. Act as the technical point of contact for external QA agencies, providing the bug documentation, test data, and feedback loops necessary to ensure our high quality standards are met at scale.
  4. Collaborate with senior engineers to develop and maintain internal testing tools and platforms that empower our entire product development organisation.
  5. Generate and manage the critical testing data required to validate complex, localized product features throughout the release cycle, ensuring high quality both at launch and through post-release updates.

Skills

Required

  • software testing
  • quality assurance
  • technically-inclined localisation role
  • testing AI features
  • LLM-powered products
  • evaluating model output
  • identify, document, and track software bugs
  • scripting language (e.g., Python, JavaScript)
  • web technologies

Nice to have

  • automated Quality Estimation (QE) frameworks
  • LLM evaluation pipelines
  • BLEU
  • COMET
  • LLM-as-judge methodologies
  • prompt engineering concepts
  • German fluency
  • Spanish fluency

What the JD emphasized

  • Hands-on experience testing AI features or LLM-powered products
  • evaluating model output for accuracy, tone, and locale-specific quality

Other signals

  • AI-powered quality frameworks
  • AI features end-to-end testing
  • LLM-generated content validation
  • automated and scale localisation testing