LLM Engineer

About the role

About Ultralytics:

At Ultralytics https://ultralytics.com/, we commit to relentless innovation in the AI space and seek team members https://www.ultralytics.com/about who resonate with our ambition to produce the world's best YOLO AI models https://ultralytics.com/yolo. If you're obsessed with AI, eager to make an impact on the world, and thrive in dynamic, high-intensity environments, we invite you to apply for a position on our team.

🌎 Where and how you can work

This full-time LLM Engineer position is based at our brand-new Ultralytics office https://www.ultralytics.com/about in London, UK, with a hybrid model of three days in-office and two days remote. Applicants must have legal authorization to work in the UK, 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 of large language models (LLMs) https://www.ultralytics.com/glossary/large-language-model-llm with our state-of-the-art computer vision https://www.ultralytics.com/glossary/computer-vision-cv technologies. Key responsibilities include:

- Developing and scaling robust LLM-powered applications using a variety of APIs, including OpenAI https://developers.openai.com/api/docs/quickstart, Anthropic https://www.anthropic.com/claude, and Gemini https://deepmind.google/models/gemini/.

- Implementing and managing multi-API workflows using tools like LiteLLM https://www.litellm.ai/ to ensure flexibility and resilience.

- Building sophisticated Retrieval-Augmented Generation (RAG) https://www.ultralytics.com/glossary/retrieval-augmented-generation-rag systems, leveraging advanced techniques like embeddings with Voyage AI https://www.voyageai.com/, rerankers, and query enrichment.

- Designing and maintaining efficient data pipelines and vector storage solutions using MongoDB https://www.mongodb.com/products/platform/atlas-vector-search Atlas Vector Search.

- Fine-tuning LLMs on custom datasets to enhance performance for specialized tasks related to our documentation https://docs.ultralytics.com/ and user support.

- Collaborating with our YOLO https://docs.ultralytics.com/models/yolo26/ development team to explore and build innovative multimodal solutions https://www.ultralytics.com/glossary/multi-modal-model.

- Automating deployment and testing processes using CI/CD pipelines https://www.ultralytics.com/glossary/continuous-integration-ci with GitHub Actions https://github.com/features/actions.

Your work will directly enhance the Ultralytics Platform https://docs.ultralytics.com/platform/ 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 and machine learning (ML) https://www.ultralytics.com/glossary/machine-learning-ml.