Note: The job is a remote job and is open to candidates in USA. Dice is a recruiting company offering a fully remote contract opportunity in a high-trust federal environment. The AI Engineer will develop scalable, secure AI solutions, including RAG and document-processing pipelines, LLM workflows, synthetic data generation, evaluation frameworks, and local model deployments, while supporting a product demonstration and future team ownership.
Responsibilities
- Develop retrieval-augmented generation (RAG) and document-processing pipelines utilizing Databricks lakehouse ingestion, OCR, chunking, embedding, and retrieval methods
- Build large language model (LLM) workflows for summarization, structured data extraction, and evidence-grounded text generation with source attribution
- Generate synthetic document corpora that meet fidelity and quality variation criteria for meaningful AI outputs
- Establish evaluation frameworks to assess retrieval quality, grounding, hallucination rates, and structured output validity, including human-in-the-loop review processes
- Package deliverables as jobs and asset bundles, track progress using MLflow, and maintain comprehensive documentation for internal team ownership
- Implement privacy-preserving synthetic data generation techniques for sensitive or restricted data sources, understanding re-identification risks
- Ensure secure local deployment of open-weight LLMs (such as TGI, Ollama, llama.cpp) in isolated or air-gapped environments
- Collaborate with cross-functional teams to integrate AI solutions within the federal environment, adhering to security and compliance standards
- Monitor and report model performance metrics, providing clear, actionable insights for evaluation and improvement
Skills
- Eligibility for a U.S. Government security clearance. In accordance with federal law, U.S. citizenship is required. (Active T5/SSBI federally adjudicated security clearance)
- Minimum of 8 years of experience in building applied ML/AI or data systems, with demonstrated delivery of personal LLM and RAG systems beyond notebook prototypes
- Extensive hands-on experience with Databricks platform, including development, deployment, and management
- Proven expertise in processing data at scale, including OCR, parsing of mixed-quality sources, and handling poor source quality
- Practical knowledge of local/self-hosted LLM serving environments such as vLLM, TGI, Ollama, llama.cpp, or similar
- Strong background in structured data extraction, grounded generation with source attribution, and model evaluation methodologies
- Experience generating privacy-preserving synthetic data from sensitive sources, with awareness of re-identification risks
- Proficiency in Python programming and related data science tools
- Prior experience working within government or defense contracting environments
Benefits
- This is a fully remote role and can be performed from any approved location within the United States.
Company Overview