Lead AI / RAG Engineer: Self-Hosted Knowledge Platform

Lead AI / RAG Engineer for a self-hosted knowledge platform at Eventum AI. Remote, $65-90/hour gross. Design and develop production RAG systems.

Responsibilities: - Design architecture and participate in hands-on development of production RAG / knowledge platform. - Develop document ingestion from Google Workspace, PDF, scans, tables, and images. - Implement permission-aware retrieval and ACL enforcement at query time. - Develop hybrid search, embeddings, reranking, and answer generation with citations. - Deploy and test open-weight models via vLLM or similar infrastructure. - Create evaluation and regression systems for retrieval, answer quality, safety, and model updates. - Responsible for production deployment, observability, CI/CD, documentation, and handover. - Work directly with Eventum AI senior team and client technical leadership. Project main tasks: - Create a production-grade self-hosted knowledge platform. - Ensure high-quality and secure retrieval of corporate data. - Implement hybrid search and answer generation with source verification. - Deploy and optimize open-weight LLM. - Build a quality evaluation and regression testing system. - Ensure reliability, observability, and production deployment of the solution.

Requirements: - 6+ years of professional experience in Software / ML Engineering. - Experience developing production LLM / RAG systems. - Strong Python and backend development skills. - Practical experience with embeddings, vector search, reranking, and document pipelines. - Experience with vLLM, SGLang, Triton, TGI, or similar technologies. - Strong cloud / infrastructure skills. - Experience with Azure, Docker, Terraform / Kubernetes, and CI/CD. - Experience with LLM evaluation and ensuring production reliability. - Willingness to work simultaneously with ML, backend, data, and infrastructure. Nice to have: - Google Workspace APIs and permissions. - pgvector / Qdrant. - BM25 + dense retrieval. - OCR / Document AI. - Multimodal models. - Prometheus / Grafana. - Experience with no-egress or security-sensitive environments.

Working conditions: - Fully remote work format. - Payment $65-90 USD gross per hour. - Work on production-grade AI / RAG systems. - Direct interaction with Eventum AI senior team and client technical leadership.