Staff Data Scientist- Causal Inference, Visual Understanding

<div class="content-intro"><p><span style="font-family: helvetica, arial, sans-serif; font-size: 12pt;">Airbnb was born in 2007 when two hosts welcomed three guests to their San Francisco home, and has since grown to over 5 million hosts who have welcomed over 2 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.</span></p></div><p><strong>The Community You Will Join:</strong></p> <p>Photos have the largest impact on our customer journey behavior and is a key lever Hosts optimize for their listings. Our visual merchandising is how Guests first encounter our supply and form impressions over our inventory options, and they are a key attribute that drives comparison and evaluation of listings, site navigation, confidence to book, and expectation setting over trip expectations. Our industry-leading photo databases provide our marketplace with listing intelligence that are a strategic asset that can help guide Hosts towards fulfilling their goals and growing their expertise at better hosting.</p> <p>You will join a team dedicated to driving innovation and powering the visual merchandising strategy over our marketplace by unlocking deep data-driven understanding and storytelling. As a member of the Customer Journey Data Science team, you will join a community of seasoned PhD practitioners to tackle frontline challenges at the core of our business, involving visual recommendations, driving host success, identifying guest decision making, causal ML, computer vision, measurement and experimentation.</p> <p> </p> <p><strong>The Difference You Will Make:</strong></p> <p>As a Staff Data Scientist, you will ensure we are deploying technologies for extracting photo insights and surfacing visual recommendations that meaningfully improve the guest and host experience. You will be responsible for building data products, inference frameworks, and intelligence to ensure that our end-to-end customer journey is effective. Your work will include:</p> <ul> <li>Specifying and estimating models to determine the behavioral response and impact of visual assets</li> <li>Delivering data and insights to enable and create impactful visual merchandising strategies</li> <li>Development and implementing experiment-based measurement and offline evaluation frameworks for computer vision and multi-modal merchandising, with a holistic assessment over the end-to-end user journey. </li> </ul> <p><strong>A Typical Day: </strong></p> <ul> <li>Prototype models: You will develop models to accurate extract visual insights, forecast the behavior of site navigation in our marketplace, and understand how users respond to our visual merchandising</li> <li>Advanced causal inference: You will explore cutting-edge causal inference techniques to quantify the impact of our photos strategy on the marketplace, informing targeted improvements and intervention.</li> <li>Cross-functional collaboration: You will collaborate closely with product, engineering, and operations to integrate data science outputs into codebases and on-the-ground workstreams aimed at improving our customer experience, driving business growth, and delivering our long-term vision strategy.<br><br></li> </ul> <p><strong>Your Expertise:</strong><strong><br></strong></p> <ul> <li>9+ years of relevant industry experience and a Master’s degree or PhD in a quantitative field</li> <li>Strong fluency in Python for hands-on IC work and advanced data analysis in SQL at scale</li> <li>Experience with causal inference and machine learning techniques, ideally in a marketplace setting. </li> <li>Preferred experience in recommendations, computer vision, information retrieval, or merchandising domains. Strong interest and agility with related AI, LLM and topic modeling tools. </li> <li>Comfort to collaborate with software engineers to understand complex systems and abstracted log ... (truncated, view full listing at source)