Required Skills & Qualifications:
- Experience with LLMs and Generative AI (OpenAI, Gemini, Claude, or open-source like LLaMA, Mistral).
- Strong understanding of SQL and data querying principles.
- Knowledge of data workflows including ETL, data validation, and data profiling.
- Proficiency in Python for interacting with LLM APIs, prompt chaining, and evaluation scripts.
- Familiarity with data modeling and analytics concepts.
- Excellent written communication for designing precise, clear prompts.
- Ability to evaluate and iterate prompts systematically.
Role Overview:
As a Prompt Engineer – Data, you will:
- Design, engineer, and optimize prompts for LLMs focused on structured and unstructured data workflows.
- Develop reusable prompt libraries for data extraction, transformation, summarization, and QA.
- Collaborate with data engineers, data scientists, and product teams to build LLM-powered data agents.
- Evaluate prompt effectiveness across accuracy, consistency, and hallucination rates.
- Continuously refine prompts using A/B testing and feedback loops to improve response quality on data-centric tasks.
Key Responsibilities:
Design high-quality prompts for:
- Query generation and validation
- Table summarization and trend extraction
- Data cleaning, normalization, and mapping tasks
- Automated data documentation and metadata extraction
- Optimize prompts for cost-efficiency across different LLM providers (OpenAI, Gemini, Claude, open source).
- Stay current on LLM advancements and integrate advanced prompting methods (chain-of-thought, tree-of-thought, self-consistency, reflection).
- Document prompt engineering guidelines and best practices for internal teams.
- Collaborate with MLOps/DevOps teams to deploy prompt-driven pipelines into productio
Job Type: Full-time
Work Location: Remote
