AI Application Engineer

Overview In this role you design and deliver production-grade LLM-powered applications for enterprise clients, focusing on RAG quality, prompt engineering, and AI safety. You will build multi-step orchestration pipelines and optimize latency, collaborating with global teams to deploy and monitor AI solutions. This is a chance to shape scalable, safe AI in production and contribute to enterprise AI delivery at a leading services provider. Compensation / Benefitscompetitive salary 130-170K USDremote or hybrid work optionsglobal collaboration opportunitiescareer growth and developmentaccess to innovation centersequitable and inclusive culture ResponsibilitiesDesign and optimize production-grade LLM-powered applicationsOwn AI quality, RAG accuracy, prompt engineering, and AI safety across multiple appsDevelop and maintain multi-step LLM orchestration pipelines using LangChain, LlamaIndex, or custom frameworksImplement and optimize RAG pipelines (chunking, embeddings, reranking, hybrid search)Design multi-turn conversational experiences with context management and session memoryIntegrate NVIDIA technologies (NIM, NeMo, NemoGuardrails, Riva) into enterprise AI appsBuild automated evaluation pipelines for model quality, hallucination detection, regression testing, and release gatingPerform latency profiling and optimization across multi-step LLM call chainsImplement AI safety guardrails (prompt injection prevention, jailbreak mitigation, topical control)Collaborate with globally distributed engineering and product teams to deliver scalable AI solutionsSupport deployment, monitoring, and continuous improvement of AI applications in production Key requirements4+ years of software engineering with at least 2 years focused on production LLM application development4+ years of Python for AI/ML application development and async programming3+ years of experience with LangChain or LlamaIndex for multi-step LLM orchestration3+ years designing and optimizing RAG pipelines and retrieval systems3+ years with vector databases, similarity search tuning, and reranking techniquesCross-functional collaborationStrong problem-solving and communicationProactive and results-drivenLangChainLlamaIndexRAG pipelines