About Halo Halo is a silently controlled language generation product that seamlessly integrates generative AI with interfaces such as electromyography (EMG) signals and eye tracking. Halo facilitates communication through an AI Persona, which consists of a textual representation of personalized, distinctive, and individualized data, tailored specifically to each user and context. By dynamically retrieving and generating information from the AI Persona in real-time, we can effectively shape the model's output through prompt engineering, ensuring the delivery of personalized and contextually appropriate responses for any given situation. Other techniques for developing a personalized large language model for each user are also considered. This product is part of the next generation of products under the Center for Responsible AI. We aim to revolutionize how people communicate, particularly those who face communication difficulties. Join us in our mission to connect these individuals with their loved ones. What's the opportunity about? We are seeking a talented and enthusiastic AI Engineer to join our Halo team, working on cutting-edge generative AI applications. This position offers a unique opportunity to join a small, dynamic, and ambitious team. You will contribute to building a product that aims to make a real difference in people's lives, while also collaborating with top research institutes like CMU, USA. Halo has already been showcased on some of the most recognized stages in the world, including: The main stage at the AI for Good Summit by the United Nations 2024, with a live demo featuring an ALS user (video here and article here ). The main stage at WebSummit 2023, where it was presented to an audience of over 8,000 people ( here ). Responsibilities Acquire knowledge, design, and implement various techniques for personalizing large language models, including prompt engineering strategies and multi-agent architectures. Conduct rigorous evaluations of AI models and assess their effectiveness. Collaborate and share knowledge with various members of the Halo team. Test and evaluate the performance of the developed solution. Requirements BS/MS degree in Computer Engineering, Computer Science, or a related field. A basic understanding of state-of-the-art AI systems, such as large language models and generative AI. Strong proficiency in Python and familiarity with data science libraries (e.g., Pandas, scikit-learn, NumPy, PyTorch, etc.). Familiarity with various generative AI and prompt engineering concepts (e.g., Chain-of-Thought, Retrieval-Augmented Generation (RAG), interactive prompting). Creative thinking and strong problem-solving skills. Fluent in English, both written and spoken. Benefits A stimulating startup environment committed to diversity and inclusion; Health insurance; Stock options; Flexible working hours; Individual budget to set up your workstation; Budget for learning and development; Other specific benefits, depending on location. Diversity Statement Unbabel is an Equal Opportunity Employer. We do not discriminate based on ethnicity, religion, colour, sex, gender identity, sexual orientation, age, or mental disability. Not only do we not discriminate, but we celebrate differences. The positions open at Unbabel are also intended for people with disabilities under DL no. 4/2019. If for whatever reason you need us to make reasonable adjustments and adaptations to our recruitment process, please let your recruiter know. (We are not accepting candidates coming from recruitment agencies at this moment)