Agentic AI Test Engineer

Comcast Comcast · Media · Mount Laurel, NJ

Seeking an AI Agentic Test Engineer to build automated evaluation frameworks using LLM-as-a-Judge patterns and maintain web/API test suites. Focus on agent evaluation and full-stack automation in Python, with experience in CI/CD and troubleshooting.

What you'd actually do

  1. Design and implement automated evaluation frameworks that utilize Large Language Models to assess the quality, accuracy, and safety of Agent responses.
  2. Perform hands-on validation of AI Agents within complex environments, moving beyond static assertions to intelligent, semantic validation.
  3. Automate the testing of real-time data streams and LLM outputs to ensure consistency and prevent regression in Agent behavior.
  4. Expert in Python for building custom test utilities, framework extensions, and automation scripts.
  5. Develop and maintain automation across TypeScript, Java, and React environments to support various application layers.

Skills

Required

  • Python
  • Agent Evaluation
  • LLM-as-a-Judge
  • Web automation
  • API automation
  • custom automation frameworks

Nice to have

  • TypeScript
  • Java
  • React
  • Streamlit
  • CI/CD pipeline management (Jenkins/Concourse)

What the JD emphasized

  • Core Language: Expert-level hands-on coding in Python.
  • AI/LLM Skills: Solid understanding of Agent Evaluation techniques and using LLMs as a judge.
  • Automation Breadth: Proven experience in Web and API automation.

Other signals

  • building automated evaluation loops for AI Agents
  • LLM-as-a-Judge patterns
  • agentic systems that are reliable, secure, and scalable