Position Overview
Builds secure, production-ready generative-AI applications using large language models, retrieval-augmented generation, vector search, prompt engineering, evaluation, and guardrails.
Key Responsibilities
- Design and implement RAG workflows, including document ingestion, chunking, embedding, vector indexing, retrieval, prompt construction, and response generation.
- Integrate LLMs and foundation models through Azure OpenAI, AWS Bedrock, Vertex AI, OpenAI-compatible APIs, or approved alternatives.
- Develop prompts, system instructions, evaluation datasets, output-quality metrics, and test automation.
- Implement guardrails for content filtering, data leakage prevention, prompt-injection resistance, access enforcement, hallucination mitigation, and harmful-output controls.
- Integrate GenAI solutions with agency data sources, repositories, workflow tools, knowledge bases, APIs, and user interfaces.
- Implement observability for model usage, latency, failures, token consumption, quality signals, security events, and user feedback.
Minimum Qualifications
4+ years in software engineering, data engineering, AI/ML, NLP, or cloud application development. Demonstrated experience with LLMs, embeddings, vector databases, RAG architecture, APIs, Python or TypeScript/JavaScript, AI security, privacy, and evaluation practices preferred.
About This Opportunity
This is a full-time remote position supporting current and upcoming JJT & Associates client work. Specific client requirements, schedules, security requirements, clearances, and other project details may vary by engagement.
JJT & Associates is committed to a professional and inclusive workplace. Employment decisions are based on qualifications, merit, business need, and the requirements of the applicable engagement.