Développeur·euse QA, Assistant IA
Core
Ensuring the quality of an integrated industrial AI assistant used by factory workers and supervisors on production floors, focusing on automated evaluation, testing, and tooling rather than manual testing.
Role type
QA specialist for an industrial AI assistant (LLM-based)
Builds
Automated evaluation suites, testing tools, and regression checks for an AI assistant
Domain
Industrial manufacturing / AI / LLMs
Deliverable
production ML models
Required skills
Python, TypeScript, test automation, LLM behavior analysis, prompt engineering, CI/CD integration, adversarial testing, exploratory testing
Preferred skills
evaluation harness experience, AI testing tools, statistics, security testing, shift-left testing, B2B platforms, Jira
Technologies
Python, TypeScript, Claude Code, promptfoo, AWS, Docker
Responsibilities
Automate QA and build tools for non-developers to add evaluations; Build and maintain evaluation suites from real user data; Define quality criteria for LLM responses using LLM-as-judge; Execute regression checks in CI for prompts, tools, and model versions; Review real conversations to identify failure trends and fix them; Contribute to adversarial testing (prompt injection, data leakage); Perform exploratory testing on new AI features and agents; Participate in feature launches to define AI-specific acceptance criteria.