Senior AI Platform Engineer — Self-Hosted LLMs, RAG & Python

We are a large organization building our own private, self-hosted AI ecosystem designed around our infrastructure, internal data, security requirements, and enterprise systems. We are looking for an experienced, hands-on AI Platform Engineer to help build the applications, services, and AI integration layer that will power this ecosystem. This is not a prompt-engineering, chatbot configuration, or AI strategy role. We need a strong software engineer who knows AI and can build production-quality systems. What You'll Build You will work with our internal technology team to: Build AI-powered applications for our users Build the backend microservices and APIs behind those applications Build an AI harness/orchestration layer connecting applications to self-hosted models Build Retrieval-Augmented Generation (RAG) services Create knowledge bases from internal organizational content Integrate AI capabilities with existing enterprise systems Implement authentication, authorization, logging, observability, and other production requirements Build reusable APIs and services that allow additional applications and models to be added over time We are not building a single chatbot. Our goal is an extensible AI ecosystem capable of supporting multiple applications, models, use cases, and internal knowledge sources. Technical Skills You should have strong experience with: Python Production APIs and microservices Self-hosted/open-source LLMs RAG architectures Embeddings and vector search Model inference APIs Containers Git-based development workflows Experience with technologies such as FastAPI, Docker, Kubernetes, vLLM, Hugging Face, LangChain, LlamaIndex, Qdrant, Milvus, pgvector, or similar technologies is valuable, but we care more about your engineering ability than experience with any particular framework. Required Qualifications 5+ years of professional software and/or AI engineering experience Strong Python development skills Demonstrated experience building production microservices and APIs Hands-on experience with self-hosted/open-source LLMs Strong understanding of RAG architectures Experience building production applications that interact with LLMs Ability to design modular, maintainable systems Strong Git and collaborative development practices Ability to document architecture, APIs, deployment requirements, and technical decisions How We Work We have an established internal technology team and work in Agile/Scrum with two-week sprints. You will collaborate directly with our engineers and technical leadership. We expect participation in sprint planning, technical discussions, demonstrations, and code reviews. You should be comfortable taking work from: requirement → architecture → code → testing → deployment → documentation We also expect you to challenge assumptions and recommend better technical approaches when appropriate. What Success Looks Like We want an architecture that allows us to build AI capabilities repeatedly rather than creating isolated AI projects. What you build should therefore be: Modular Secure API-driven Well documented Observable Maintainable by our internal team Capable of supporting multiple applications and models Designed to minimize unnecessary dependence on proprietary AI platforms Knowledge transfer is important. Our internal engineers need to understand and maintain what we build together. We're looking for someone who has built these systems, not someone whose experience is primarily prompt engineering, AI consulting, or connecting applications to commercial AI APIs. There is potential for a longer-term engagement as the AI ecosystem expands.