Expert RAG Engineer

We are looking for an Expert RAG Engineer to design, build, and own the retrieval-augmented generation systems behind our AI products. Your core work will be architecting production-grade RAG pipelines end to end: ingestion, chunking, embeddings, hybrid retrieval, reranking, evaluation, and serving on top of open-source LLMs (Ollama, llama.cpp) in cloud, on-prem, and air-gapped environments.

This is a senior, hands-on role. You’ll define RAG architecture standards, make the key technical decisions on retrieval quality, latency, and cost, and provide technical guidance to AI Developers on the team. You’ll work closely with engineering, product, and client stakeholders to turn complex knowledge sources into reliable, measurable, and secure AI features.

Key Responsibilities

RAG Architecture & Engineering (core focus)

• Design and own end-to-end RAG architectures, from document ingestion to grounded, cited responses.
• Build advanced retrieval strategies: hybrid search (dense + BM25/sparse), reranking (cross-encoders), query rewriting/expansion, metadata filtering, and multi-step or agentic retrieval.
• Define chunking and indexing strategies for diverse sources, including structured and semi-structured data (XML/XSD, Office formats, PDFs, scanned documents).
• Select, benchmark, and fine-tune embedding models; manage vector databases (pgvector, Qdrant, Weaviate, Milvus, Azure AI Search, or similar) at scale.
• Build evaluation frameworks for retrieval and generation quality (View email address on aiapply.co, MRR, nDCG, faithfulness, answer relevance) and run them continuously as part of CI/CD.
• Reduce hallucinations through grounding, citation, guardrails, and confidence handling.
• Implement permission-aware and multi-tenant retrieval so users only access data they’re authorized to see.
• Develop MCP (Model Context Protocol) integrations that expose RAG systems and internal data sources to LLMs and agents.
• Optimize LLM serving with Ollama and llama.cpp: quantization (GGUF), GPU/CUDA tuning, batching, caching, and offline/air-gapped deployments.

Technical Leadership

• Set RAG architecture standards, reference implementations, and best practices for the AI department.
• Review designs and code, and provide technical guidance to AI Developers.
• Translate client and product requirements into retrieval strategies with clear quality, latency, and cost targets.
• Evaluate emerging techniques (GraphRAG, long-context vs. retrieval trade-offs, agentic RAG) and recommend what’s worth adopting.

DevOps & Infrastructure (supporting)

• Deploy and scale RAG services on Microsoft Azure and AWS using Docker and Kubernetes.
• Maintain CI/CD pipelines for AI services (Azure DevOps preferred; GitHub Actions / GitLab CI also used).
• Build observability for RAG systems: tracing, retrieval diagnostics, latency, cost, and quality dashboards.
• Support event-driven ingestion and re-indexing pipelines using Apache Kafka.
• Troubleshoot across the stack - dig into root causes rather than patching symptoms.

Requirements

• Experience & Core RAG Expertise: 6+ years in backend, ML, or infrastructure with 2+ years building production RAG systems, including end-to-end architecture, hybrid retrieval, reranking, chunking, indexing, and evaluation frameworks.
• Python & Vector Databases: Expert-level Python programming combined with deep knowledge of embeddings, vector databases, and metrics-driven evaluation.
• LLM Optimization & Serving: Hands-on experience serving and optimizing open-source LLMs (Ollama, llama.cpp) with quantization, alongside GPU/CUDA troubleshooting and performance tuning.
• DevOps, Infrastructure & MCP: Strong proficiency in Microsoft Azure, AWS, Docker, Kubernetes, Linux, CI/CD pipelines (Azure DevOps preferred), and Model Context Protocol (MCP).
• Leadership & Collaboration: Strong communication, technical leadership, and ownership to define architectural standards and collaborate effectively in a distributed, remote environment.