Staff Data Scientist

Airbnb was born in 2007 when two Hosts welcomed three guests to their San Francisco home, and has since grown to over 4 million Hosts who have welcomed more than 1 billion guest arrivals in almost every country across the globe. Every day, Hosts offer unique stays and experiences that make it possible for guests to connect with communities in a more authentic way. The Community You Will Join: Airbnb is a mission-driven company dedicated to helping create a world where anyone can belong anywhere. Customer Support (CS) aims to build the world's most loyal travel community through exceptional service. And AI has been and will continue to be the key to unlock timely, reliable, and empathetic solutions at scale. As a Data Scientist working on Algorithms, CS, you will have the opportunity to collaborate with a strong team of engineers, product managers, designers and operation agents to build scalable and robust systems to detect, mediate, and solve issues across a wide spectrum. You will be able to create meaningful impact through deep scientific understanding and interventions for Airbnb CS's most critical challenges. The Difference You Will Make: For this role, we're seeking an NLP expert to join the Customer Support Data Science team, and have a direct opportunity to contribute, influence, and lead in the CS x AI space by designing scalable scientific solutions for problems like: - Detect intents across various entry points with large scale unstructured data using NLP methods - Generate most helpful self-solve contents in line with Airbnb policy leveraging generative AI to automate low-complexity issues - Route high-complexity issues to agents with personalized strategy, incorporating user history and agent performance - Summarize customer journey and conversations with foundational or fine tuned large language models (LLM) to assist CS agents - Collect evidence from images and receipts to gauge monetary value of the request leveraging computer vision A Typical Day: - Identify high impact business opportunities through data exploration and model prototype, translate business problem into scientific formulations - Work collaboratively with cross functional partners including software engineers, product managers, operations and research, to refine requirements for machine learning models, drive scientific decisions, and quantify impact - Hands-on develop, productionize, and operate machine learning models and pipelines at scale, including both batch and real-time use cases, structured and unstructured data - Build reusable, high-performing, scalable machine learning models with internal paved path tooling, incorporating third-party information and state-of-the-art innovations - Regularly present work internally at monthly meetings to technical, engineering and product stakeholders to iterate and generate excitement on roadmap progress - Publish externally and engage with the scientific community to advance Airbnb's standing - Your Expertise: - 9+ years of relevant industry experience (e.g. ML scientist, tech lead, junior faculty) and a Master's degree or PhD in relevant fields - Strong fluency in Python and SQL, experience with Tensorflow, PyTorch, Airflow and data warehouse - Deep understanding of Machine Learning lifecycle best practices (eg. training/serving, feature engineering, feature/model selection, labeling, A/B test), algorithms (eg. gradient boosted trees, neural networks/deep learning, optimization) and domains (eg. natural language processing, computer vision, personalization and recommendation, anomaly detection) - Proficiency with LLMs and/or related AI, NLP, CV, UGC/content understanding topics including deep learning, information retrieval, or knowledge extraction. For example, BERT, GPT-2/3/4, LLaMA, Mistral - Comfort to collaborate with software engineers to understand complex systems and abstracted logs - Proven ability to communicate clearly and effectively to audiences of varying technical levels - Proven mix of strong intellectual curiosity with high level of pragmatism and engagement with the technical community. Publications or presentations in recognized journals/conferences is a plus - Ability to take a product-oriented mindset in using conceptual and innovative thinking to develop and apply solutions taking into consideration the user experience