LLM Engineer

Who We Are AtUltralytics , we relentlessly drive innovation in AI, building the world's leadingYOLO models . We're looking for passionate individuals obsessed with AI, eager to make a global impact, and ready to excel in a dynamic, high-energy environment. Joinour teamand help shape the future ofVision AI .Location and Legalities This full-time LLM Engineer position is based onsite in ourUltralytics officein Madrid, ES. Applicants must have legal authorization to work in the Spain, as Ultralytics does not provide visa sponsorship.What You'll Do As an LLM Engineer at Ultralytics, you will spearhead the development and integration of advanced language model capabilities across our ecosystem. You will work on cutting-edge projects that merge the power ofLarge Language Models (LLMs)with our state-of-the-artcomputer visiontechnologies. Key responsibilities include: Developing and scaling robust LLM-powered applications using a variety of APIs, includingOpenAI ,Anthropic , andGemini . Implementing and managing multi-API workflows using tools likeLiteLLMto ensure flexibility and resilience. Building sophisticatedRetrieval-Augmented Generation (RAG)systems, leveraging advanced techniques likeembeddingswithVoyage AI ,rerankers , andquery enrichment . Designing and maintaining efficient data pipelines and vector storage solutions usingMongoDBAtlas Vector Search. Fine-tuning LLMs on custom datasets to enhance performance for specialized tasks related to ourdocumentationand user support. Collaborating with ourYOLOdevelopment team to explore and build innovativemulti-modal solutions . Automating deployment and testing processes usingCI/CD pipelineswithGitHub Actions . Your work will directly enhance theUltralytics HUBplatform and empower our global community of developers and researchers.️ Skills and Experience 5+ years of experience in software engineering, with a strong focus on AI andMachine Learning (ML) . Expert-level proficiency inPythonand deep learning frameworks, particularlyPyTorch . Proven experience building and deploying applications with LLM APIs such asOpenAI ,Anthropic ,Gemini , andDeepSeek . Hands-on experience with the full RAG pipeline, including vectorembeddings ,rerankers , and data indexing in databases likeMongoDB . Practical knowledge ofLLM fine-tuning , prompt engineering, and performance optimization. Familiarity with MLOps principles and tools, includingCI/CDwithGitHub Actions . A strong interest incomputer visionand an understanding of object detection models likeUltralytics YOLO11is a significant plus. Excellent problem-solving skills and the ability to perform in a fast-paced, high-intensity environment.Cultural Fit - Intensity Required Ultralytics is a high-performance environment for world-class talent obsessed with achieving extraordinary results. We operate at a relentless pace, demanding exceptional dedication and an unwavering commitment to excellence, guided by ourmission, vision, and values . Ourteamthrives on audacious goals and absolute ownership. This is not a conventional workplace. If your priority is predictable comfort or a standard work-life balance over the relentless pursuit of progress, Ultralytics is not for you. We seek driven individuals prepared for the profound personal investment required to make a defining contribution to the future of AI.Compensation and Benefits Competitive Salary:Highly competitive based on experience. Startup Equity:Participate directly in our company's growth and success. Hybrid Flexibility:3 days per week in our brand-new office - 2 days remote. Generous Time Off:24 days vacation, your birthday off, plus local holidays. Flexible Hours:Tailor your working hours to suit your productivity. Tech:Engage with cutting-edgeAI projectsand contribute to leadingopen-sourcesoftware. Gear:Brand-new Apple MacBook and Apple Display provided. Team:Become part of a supportive and passionateteam environment .If you are driven to redefine the capabilities of machine learning and eager to make a significant impact,Ultralyticsoffers an exceptional career opportunity.