Lead AI Engineer
Summary
Lead the design and deployment of AI-powered search and document processing systems using Azure AI Search, RAG, and Kubernetes, optimizing retrieval and relevance for EPRI’s technical content.
Job Title:
Lead AI EngineerLocation:
EPRI Charlotte OfficeJob Summary and Description:
The Lead AI Engineer is to design, implement, and support AI-driven document processing, retrieval, and search solutions using Azure AI Search, Retrieval-Augmented Generation (RAG), Knowledge Graphs, Query Orchestration, Prompt Engineering, and Kubernetes-based container deployments.
Key Responsibilities
- Architect & Deploy AI Solutions: Build AI-driven search and document intelligence systems using Azure AI Search, Knowledge Graphs, and RAG techniques.
- Query Orchestration: Develop strategies to route and structure user queries efficiently across multiple retrieval systems.
- RAG-Based Applications: Implement and fine-tune applications for intelligent knowledge retrieval from structured and unstructured documents.
- Containerized Deployments: Deploy and manage AI applications using Azure Kubernetes Service (AKS) for scalability.
- Vector Search Optimization: Enhance document retrieval through optimized embeddings and hybrid search techniques.
- Open-Source Integration: Utilize tools like Tesseract OCR, PyMuPDF, and Pillow for document processing.
- API Integration: Connect with Profile APIs, Product Metadata, and Downloads to enrich indexing and search capabilities.
- Compliance & Security: Ensure adherence to export control restrictions and secure document handling best practices.
- Monitoring & Optimization: Troubleshoot and optimize AI-based workflows for performance and reliability.
- Stakeholder Collaboration: Work closely with business and technical teams to refine AI-powered document solutions.
Required Skills & Experience
- Bachelors or Masters Degree in Computer Science or related areas, applicable professional certification with 7+ years of progressive experience providing solutions to complex program/system problems in a business environment in
- 5+ years in AI/ML, cloud-based search, and document processing.
- Expertise in Query Orchestration for complex AI pipelines.
- Strong knowledge of RAG architectures for AI-powered search.
- Hands-on experience with Azure AI Search, Document Intelligence, and Cognitive Services.
- Proficiency in vector search, embeddings, and hybrid retrieval techniques.
- Experience with Kubernetes (AKS) and containerized deployments.
- Familiarity with Tesseract OCR, PyMuPDF, and Pillow.
- Strong Python development skills for AI pipelines.
Specialized Expertise
Search & RAG
- Semantic, BM25, similar ranking and vector search optimization.
- Custom scoring profiles and relevance tuning.
- Evaluation metrics (nDCG, MRR, precision@k).
- Query rewriting, synonym maps, and semantic expansion.
- Integration with RAG and LLM pipelines for optimized context retrieval.
Prompt Engineering
- Systematic prompt design and evaluation.
- RAG-oriented prompting with grounding and guardrails.
- Instruction hierarchies and multi-agent orchestration.
- Domain adaptation for EPRI’s technical language.
- Continuous improvement through telemetry and quality analysis.
Preferred Qualifications
- Experience with hybrid cloud AI solutions (on-prem + cloud).
- Familiarity with Azure OpenAI, LangChain, or AI Foundry.
- Deep knowledge of multi-index query orchestration.
- Expertise in Azure AI search semantic and vector profiling.
- Expertise in vector other databases such as (FAISS, Weaviate, Pinecone).
- Background in NLP, document classification, and entity extraction.
- Understanding of export control compliance and secure document handling.