Applied AI Data Scientist / Prompt Engineer

Summary:
Develop analytics and AI solutions to transform raw data into meaningful insights using statistics, machine learning, and visualization software, with a strong focus on LLMs and generative AI in a public-sector context.

Responsibilities

• Collect, process, and analyze structured and unstructured data using data mining, modeling, NLP, and ML techniques.
• Develop predictive models and algorithms and design automated data pipelines and workflows.
• Build dashboards, reports, and visualizations; collaborate with multiple teams to refine requirements.
• Implement AI governance and safety guardrails to reduce hallucinations, bias, and security risks (e.g., prompt injection).
• Develop LLM evaluation benchmarks with automated metrics and human-in-the-loop feedback.
• Identify parameter-efficient fine-tuning (PEFT/LoRA) opportunities for state government datasets.

Requirements

• Minimum 3 years of data science experience.
• Strong background in statistical analysis and ML; proficiency in SQL, Python, R, or similar.
• Experience with ML libraries/frameworks and methods such as regression, clustering, and classification.
• 2+ years hands-on with models such as GPT-4, Claude, Llama, Gemini or similar, and their APIs.
• Expert-level with orchestration tools such as LangChain, LlamaIndex, or Haystack.
• Experience with vector databases (Pinecone, Weaviate, Milvus, pgvector) and synthetic / instruction-tuning datasets.
• Preferred: Experience in regulated/public-sector environments with PII/PHI and ethical AI standards.

Education

• Master’s or PhD in computer science, statistics, mathematics, economics, or related field.
• Three years of equivalent related experience may substitute for a Bachelor’s degree.