Conversational AI Engineer, System Prompt

Responsibilities

• Architect core system prompts and manage context windows to ensure responsive, contextually relevant, and logically sound AI reasoning.
• Build prompt systems that maintain a consistent, empathetic personality and weave long-term user memories into real-time dialogue.
• Design advanced prompt chains and fallback logic to handle conversational tangents, interruptions, and complex emotional states.
• Implement prompt-driven logic for multi-agent frameworks to act as a routing engine between users, APIs, and internal engines.
• Develop custom evaluation frameworks to measure conversational dynamism, warmth, engagement, and hallucination reduction.

Requirements

• 2-3 years of experience in Conversational AI/NLP, specifically designing and deploying complex LLM applications in production.
• Deep familiarity with state-of-the-art prompt engineering techniques including Few-Shot, Chain-of-Thought, and ReAct.
• Experience building agentic and RAG architectures using frameworks such as LangGraph, LlamaIndex, or Haystack.
• Proficiency with programmatic prompt optimization tools like DSPy.
• Ability to optimize prompt systems for scale, API cost, and latency.

About the Company

Known is a matchmaker that uses voice-led AI to support users like a friend. Our mission is to empower humanity by applying general intelligence to human connection. Our engineering team includes creators of widely used AI-driven consumer products such as Uber Eats, Uber, Faire, and Afterpay. We are based in San Francisco and work with a uniquely intimate multi-modal data set to power real-time, personalized conversations.