Artificial Intelligence Engineer, Runna

• Designing and building AI powered systems - you have solid experience working with LLMs in production
• Testing your code to ensure high quality solutions that are maintainable and scalable
• Analytical and detail-oriented, with a commitment to producing high-quality work
• A pragmatic mindset, with excellent communication and collaboration skills
• Proficiency in Python (object orientated) programming, with experience writing production quality code
• Enthusiasm for our ways of working which include
• Working within a highly-skilled engineering team in a fast-paced, iterative environment
• Iterative development, continuous deployment and test automation
• Knowledge sharing, pair programming, collaborative design & development
• Shared code ownership & cross-functional teams

Desirable

• As a cross functional (engineering, product, design, coaching) team of >20, the Train Team work at the heart of Runna’s mission: delivering the best training possible to hundreds of thousands of runners worldwide.
• From creating and dynamically adjusting optimal, personalised training plans at scale, to analysing performance data to make training adjustment recommendations, to providing insight to users derived from their training and environment - we do it all
• You’ll be building AI-powered features that leverage large language models to transform the running experience for users around the world.
• As part of this work, you’ll be working closely with the engineering, product and coaching teams to build components that use LLMs to interpret, generate and personalise training insights - whether through conversational interfaces, knowledge-based reasoning or dynamic content generation. You’ll be part of the Train team here and we’ll all support you along this exciting journey!
• Building, testing and delivering new AI-driven features that use LLMs to enhance our personalised training experience, including conversational coaching, dynamic feedback, and content generation for hundreds of thousands of active users
• Working across the full stack of applied LLM engineering - from prompt design and model evaluation, to orchestration, deployment and iteration
• Keeping up to date with developments in the AI space, frontier models, new releases and entrants etc and ensuring we (the whole of engineering) are best utilising the latest approaches, models and strategies
• Designing and refining prompt strategies and retrieval-augmented generation (RAG) pipelines to ensure grounded, accurate and context-aware outputs
• Collaborating with coaches to help translate deep training knowledge into scalable, intelligent systems powered by AI
• Applying a data-driven and experimental approach to model development, evaluation and continuous improvement of LLM-based features - such as designing and implementing robust evaluation frameworks to measure LLM output quality, factuality, usefulness, and safety - both via automation and human feedback
• Experimenting (e.g. A/B tests in production) in order to ensure high quality outcomes
• Writing clean, well-tested Python code and contributing to the backend systems that support model orchestration, inference and integration into user-facing features

• Introductory chat (45 minute video call - intros; chat about experience; live coding)
• Take Home Technical Task (~1-2 hours to complete)
• Technical interview (75 minutes - the first half of the call will be used to discuss the take-home technical task from the previous stage and the second half will be some general architecture/tech questions)
• Meet the team and in-person chat (in-person chat with different people in the team and some technical discussion)