arbiter-ai - QA Engineer, Voice AI Agents

• As a QA Engineer reporting to the Head of Product, you will be the person who digs into calls and platform issues. You will review conversations, trace failures through transcripts, logs, and campaign configuration, find the root cause, and work with engineering and deployment teams to fix it. Work crossfunctinally with engineers, deployment managers, and clinical and nurse operations staff. This role combines hands-on investigation, structured QA process, and growing automation. • Call and issue debugging: • Investigate reported and detected issues on live campaigns, including dropped calls, wrong agent responses, failed handoffs, incorrect scheduling outcomes, and data mismatches. • Trace problems across call recordings, transcripts, system logs, campaign configuration, and integration data to find the actual root cause. • Write clear, reproducible bug reports that engineers can act on right away, with severity, member impact, and supporting evidence. • Triage incoming issues, separating agent behavior from configuration errors, data problems, and platform bugs, and route each to the right owner. • Run pre-launch QA on new campaigns and workflow changes: test scripts, edge cases, escalation paths, and handoffs to nurses or live staff. • Carry out ongoing conversation QA on live campaigns against defined quality standards, and flag trends before they turn into customer issues. • Check that campaigns behave as designed: correct targeting, sequencing, disposition logic, and outcome capture. • Process and automation: • Help build and maintain the QA framework: test cases, scoring rubrics, regression suites, and release checklists. • Automate repetitive checks where it makes sense, such as transcript analysis, config validation, and outcome reconciliation. • Track quality metrics and report patterns to deployment, product, and engineering so issues get fixed at the source.