Machine Learning Engineer

At Deliveroo, we have a premier data science department, intent on enabling superior human and machine decision-making. We operate across the entire company - in product, operational and platform teams, solving some of the most intriguing problems. We are fast-paced, always on the hunt for fresh ideas, and we maintain complete transparency regarding the decisions we make and their underlying reasons. There's a plethora of queries we need to explore and many we haven't yet faced. How can data and technology aid restaurants to thrive as consumer tendencies shift? How can we anticipate what someone will order for dinner before it even crosses their mind? These are just some of the complex issues we're tackling at Deliveroo - and there's no challenge that couldn't be yours. No solution is tied to a specific team, which means opportunities for growth and personal impact are vast. Our team employs a full range of data science skills: constructing analytical tools; informing decision-making across the business through custom and automated analysis; conducting experiments; undertaking causal analysis; influencing planning and prioritization with solid impact estimates; developing production-ready machine learning and optimization models; and enhancing the entire company's understanding of data literacy and data-driven decision-making. Our data scientists originate from numerous fields including mathematics, statistics, economics/econometrics, physics, psychology, finance, computer science, operations research, cognitive neuroscience, and engineering. Some have formal data science training while others do not. Our team is comprised of individuals from around the globe, and we even have a designated data science diversity committee. At Deliveroo, data scientists report to our data science management team. We have a vibrant, active data science community, with a robust technical review process, a career advancement framework, and plenty of learning opportunities. About the Role In this role, you'll work on metadata generation for input into recommendation engines and other algorithm-driven decision-making. You'll excel in complex prompt engineering and focus significantly on developing proprietary models using techniques like fine-tuning and distillation. Proficiency in deep learning and NLP is essential, and you can expect to use frameworks such as Pytorch and Huggingface in your daily work. You'll work with a cross-functional team of engineers, analytics and inference specialized data scientists, and product managers to develop systems that make automated decisions on a massive scale. We evaluate all our decision-making mechanisms through robust experimentation, powered by our world-class experimentation platform. Requirements: - A solid understanding of machine learning and/or operations research fundamentals and knowledge of their application - Experience with deep learning and NLP, particularly with Pytorch or other deep learning frameworks - Experience with productionizing ML models and constructing ML pipelines (experience with Huggingface is a plus) - Proficiency in exploratory data analysis, modeling/algorithm prototyping, and selection, as well as model pipeline design and development. Experience with deploying models to production and monitoring experiments is a plus - Excellent communication skills - verbally, in writing, in coding, and code-review, both to technical and non-technical audiences - Comfortable in a rapidly evolving environment with shifting requirements - A practical mindset, leaning towards simplicity, with a focus on generating impact Some Nice-to-haves: - Experience with image-based deep learning or multimodal models - Multilingual natural language processing experience in French, Italian, Arabic, Cantonese - AWS development experience Benefits and Diversity At Deliveroo, we understand that our people are the core of the business and we prioritize their well-being. We offer an expansive range of competitive benefits in sectors including health, family, finance, community, convenience, growth, time away, and relocation. We believe a great workplace is a reflection of the world and its wonderful diversity. As such, we do not judge anyone based on their gender, race, sexuality, religion, or a secret disdain for coriander. All you need is a passion for (most) food and the aspiration to be part of one of the most exciting tech companies' scale-up.