QA Engineer (Agentic Memory Testing)

Quick Facts

• Role: QA Engineer (Agentic Memory Testing)

Description

You will validate agent memory capabilities within an enterprise AI Agent Platform being built from the ground up. The focus is ensuring AI agents retain, retrieve, and manage contextual information reliably across individual conversations and multiple sessions. You’ll test stateful behaviors, memory extraction and retrieval, lifecycle controls, and regression scenarios to improve performance without introducing incorrect or irrelevant context.

Responsibilities

• Design and execute functional tests for short-term and long-term agent memory

• Validate in-session context continuity and cross-session memory recall

• Test memory extraction strategies to verify summaries, facts, and preferences are captured correctly

• Validate memory retrieval to ensure relevant context is returned while filtering irrelevant information

• Test memory lifecycle behaviors, including TTL expiration, deletion, and namespace isolation

• Develop Python-based automated tests for agent memory workflows and API-driven services

• Perform regression testing to ensure memory changes do not negatively affect agent response quality

• Validate stateful agent behaviors including context retention and personalization

• Test memory isolation to prevent inappropriate context sharing between users, sessions, or agents

• Validate user deletion and privacy-related memory workflows

• Integrate memory testing into broader agent quality and regression suites

• Collaborate with AI and platform engineers to identify memory-related defects and improve agent reliability

Requirements

• 4+ years QA engineering for AI/ML or conversational AI systems

• Experience testing stateful agent behaviors (context retention, personalization)

• Functional test automation for API-driven services

• Regression testing methodologies

• Agent memory functional testing (short-term and long-term)

• Memory extraction validation (summary/facts/preferences)

• Memory retrieval testing (relevant retrieval and irrelevant filtering)

• Memory lifecycle testing (TTL expiration, user deletion, namespace isolation)

• Agent regression testing to ensure memory does not degrade agent response quality

• Python test automation for memory workflows