Software Engineer, Machine Learning, Trust
What is Trust at Airbnb?
Over two million people stay on Airbnb every night and the Trust Engineering team keeps our hosts and guests safe and supported throughout the entire Airbnb experience.
As part of the Trust Engineering team you will be in charge of designing and building scalable and robust systems to detect and mitigate fraud across our entire platform. You will be deeply involved in the technical details of building highly available and real-time risk detection services in close collaboration with product, data science and operations teams to understand ever evolving attack vectors and to make Airbnb the most safe and trusted community.
What is Machine Learning Engineer on Trust at Airbnb?
This team is responsible for identifying and taking an action on suspicious accounts and logins, objectionable content, fraudulent transactions. We use machine learning to examine photos, videos, text, and anything else to decide if it contains nudity, pornography, violence, or hate speech. We make use of computer vision, machine learning, graph analysis, and large scale systems programming to ensure that the content on Airbnb is safe, legitimate and inclusive. The ideal candidate will have industry experience working on a range of classification problems, machine learning, deep learning, and natural language processing.
As a Machine Learning Engineer on Trust, you help keep Airbnb safe by protecting Airbnb and its community from account takeovers, objectionable content, fraudulent transactions, hate speech, spam and other kinds of fraud and abuse. You are in charge of applying machine learning, computer vision, natural language processing to examine logins, accounts, messages, photos, videos, text, and anything else to decide if is legitimate. You contribute to a variety of critical organizational efforts like:
- Building shared infrastructure to enhance and automate training and production scoring pipelines (across streaming and batch layers)
- Working on feature exploration and feature engineering to increase models' precision and recall
- Investing in sequence modeling, natural language processing, computer vision to analyze event, state, temporal and graph features to catch fraud and abuse on the platform
- We built several Random forest and XGBoost-based machine learning models to classify chargebacks, account takeovers, fake inventory, offline scams, property damage and more.
- The team built a shared infrastructure to support hierarchical modeling and low-latency feature extraction (streaming).
- We employed the unsupervised approach to get low-dimensional sequence representations, construct features for downstream fraud models and form clusters to gain insights into fraud trends.
- 5+ years of industry experience or a PhD + relevant industry experience
- Bachelor’s and/or Master’s degree, preferably in CS, or equivalent experience
- Ability to write high performance production quality code.
- Industry experience building and productionizing innovative end-to-end Machine Learning systems
- Good understanding of common families of models, feature engineering, feature selection and other practical machine learning issues, such as overfitting
- Experience working on classification problems is a plus, e.g. payments fraud, text/sentiment classification, or spam detection
- Machine learning, deep learning, natural language processing, computer vision, pattern recognition, large-scale data mining or artificial intelligence experience is a plus, but not required
- Competitive salaries
- Quarterly employee travel coupon
- Paid time off
- Medical, dental, & vision insurance
- Life insurance and disability benefits
- Fitness discounts
- Flexible Spending Accounts
- Apple equipment
- Commuter subsidies
- Community involvement (4 hours per month to give back to the community)
- Company sponsored tech talks and happy hours
- Breakfast, lunch, and dinner
- Much more...
Airbnb has Engineering offices along the West Coast in the U.S. and our teams are growing quickly! Contact us to find out which office works best for you.
Back to top