Key Responsibilities:

• 
  Design and iterate prompts and prompt chains for large language models (GPT-4, Claude, Mistral, etc.)

• 
  Collaborate with product and tech teams to develop emotionally intelligent, clinically sound interactions

• 
  Evaluate LLM responses for accuracy, tone, safety, and customer experience using metrics, product feedback, and business goals..

• 
  Prototype intelligent conversational flows and decision trees for both customers and contact center agents.

• 
  Use retrieval-augmented generation (RAG), vector databases, and feedback loops to improve model performance

• 
  Help document best practices, prompt libraries, and internal tools for scalable deployment

Requirements

Skills and Qualifications:

• 
  Strong understanding of how large language models work and how to guide them through effective prompting

• 
  Proficiency in Python (3+ years) and experience working with LLM APIs (OpenAI, Anthropic, Hugging Face, etc.)

• 
  Experience with LangChain, vector databases (like Pinecone or Weaviate), and RAG pipelines

• 
  Comfort working with structured data formats (JSON, YAML) and REST APIs in building testable prompt frameworks.

• 
  Ability to think both creatively (language, UX) and analytically (model behavior, logic)

• 
  Bonus: Exposure to AI voicebots + chatbots