LLM Researcher

Role overview An applied LLM researcher role building intelligent, automated systems that enhance customer-facing applications in the sports and entertainment space. The work contributes directly to dynamic, real-time content such as live commentary for games like FIFA or NBA, plus interactive avatar experiences, combining reasoning, tool use, and creative generation.

Responsibilities - Select, fine-tune, or train language models adapted to specific use cases and turn successful prototypes into reliable, maintainable, and observable production services - Build retrieval-augmented generation systems using vector or graph databases and integrate them into the broader product - Deploy, monitor, and integrate model-backed services into the overall architecture - Apply function calling, structured output generation, and agentic reasoning workflows to production systems - Collaborate closely with a small, fast-moving team of ML engineers, product developers, and domain experts

Requirements - Solid hands-on experience with LLMs in production, including training, fine-tuning, inference, and tool integration - Practical experience building RAG-based systems and working with vector or graph databases - Strong understanding of function calling, structured output generation, and agentic reasoning workflows - Proficiency in Python and key libraries such as PyTorch, Hugging Face Transformers, and FastAPI

Nice to have - Experience with sports broadcasting or data-driven content generation such as live commentary and analytics - Familiarity with multimodal LLMs that combine textual and visual inputs - Experience with efficient model serving tools such as vLLM, Triton, or DeepSpeed - Contributions to open-source AI tools or research publications in NLP, multimodal AI, or agent systems

Benefits and work setup - Remote-first role with the option to join a Prague office, working with cutting-edge LLM, RAG, and multimodal frameworks