Conversational AI & Prompt Engineer

Sourced from jobneticx.mydiscussion.net

• Analyze conversation transcripts and user feedback to identify areas of confusion, failure, and prompt leakage.
• Define and track conversational KPIs (resolution rate, containment rate, user satisfaction) with the Customer Impact Team.
• Optimize prompts and model selection for cost efficiency, response latency, and scalability in production environments.
• Design and maintain evaluation frameworks (golden datasets, LLM-as-judge, rubric-based scoring) to measure prompt performance.
• Implement guardrails to reduce hallucinations, prevent prompt injection, and ensure compliant, safe responses.
• Design, map, and implement complex conversation flows, including error recovery and contextual handoffs to human support.
• Own continuous optimization of system prompts and instructions for LLMs (Gemini, OpenAI).
• Design and optimize structured outputs, function calling, and tool-routing logic for accurate data capture and integrations.