Prompt Engineer

This isn't just about writing instructions; it's about ensuring quality, truth, and safety in every AI interaction. You'll be a champion of best practices, documenting the magic so we can scale these incredible experiences across all of Cashea.

If you're fascinated by the frontier of AI and have a knack for turning complex ideas into elegant instructions, you'll be at the forefront of innovation here. You'll get to dive into the exciting world of prompt tuning, fine-tuning, and RAG, constantly pushing the boundaries of what's possible.

• Design, iterate on, and optimize complex prompts for language models like GPT-4/Claude/Gemini.

• Work closely with the AI/ML and product teams to create coherent, safe, and useful conversational experiences.

• Evaluate model performance based on specific metrics (fluency, accuracy, alignment with business objectives).

• Ensure the quality, factual accuracy, and safety of AI-generated responses.

• Document frameworks and best practices to scale the use of prompts across different Cashea products.

• Participate in the research and experimentation with new techniques in prompt tuning, fine-tuning, and RAG (retrieval augmented generation).

• +2 years of experience in designing and optimizing prompts for LLMs (e.g., GPT-3.5/4, Claude, PaLM, etc.).

• Deep understanding of how language models work, including zero-shot, few-shot, chain-of-thought, and tool usage.

• Experience with tools such as LangChain, LlamaIndex, Weaviate, Pinecone, or similar.

• Analytical skills to measure the impact of changes in prompts.Advanced level of writing and comprehension in both English and Spanish.

• Previous experience working with product, UX, or development teams in agile environments.

401K
paid time off