AI Engineer

As an AI Engineer, you'll be a vital part of our team, working together to build Donna, our perfect AI-sales assistant for field sales reps. Donna empowers them to focus on what they do best - selling. Expect to navigate a wide range of challenges including developing a scalable LLM framework, fine-tuning and personalising prompt engineering, orchestrating multiple AI agents, and implementing advanced RAG systems. Your goal will be to craft the optimal user experience with our LLMs. You’ll join our ambitious Donna team as one of the first hires. This means you’ll have the opportunity to be part of an early-stage start-up and to learn from experienced serial entrepreneurs. If you're ready to embark on an exciting journey from the early days and make a significant impact in a growing team with global ambitions, Donna is the place for you! What you’ll be doing A non-exhaustive list of features we could use your help with: Leveraging best practices and industry-leading tools and technologies to solve real-life problems with LLM solutions. Fine-tuning and optimising prompting techniques (like prompt-chaining) to improve the accuracy and relevance of responses for our users. Integrating and synthesising a massive amount of contextual data (CRM data, call recordings, emails) to enrich the LLM context and provide more accurate and personalised outputs. Developing algorithms in multilingual conversational systems. Developing and enhancing RAG systems to ensure high-quality and contextually appropriate responses. Building out an MLOps and ML Observability suite to ensure the output of our models maintains a high standard of quality, accuracy, and relevance. We believe in shipping fast to gather feedback quickly from our customers. We like to keep things simple, so we leverage as much as possible the services offered by key vendors. You’ll report directly to our co-founder & CTO, but you’ll work together with other departments (and customers) in detailing the requirements of features and make trade-offs together in order to maximise value while minimising complexity.