Applied Scientist, Trust & Safety

About the Applied Science Team

The Applied Science team for Trust and Safety plays a critical role in maintaining a healthy and secure marketplace by leveraging ML to detect and prevent fraudulent activities and policy violations. Our work directly contributes to improving the platform experience for both customers and professionals by ensuring high-quality connections and safeguards. We oversee a variety of problems spanning machine learning, statistics, computer science and more specifically, detecting and preventing fraud to improve marketplace trust and safety.

The challenge 

Our team is at the forefront of a massive amount of user data related to interactions and activities on the Thumbtack platform. We need to think a lot about scaling our detection capabilities as well as how to provide robust security and trust for a wide set of users and scenarios. We are constantly working to improve our ability to identify and mitigate risks such as fraudulent accounts, spam messaging, and payment fraud, ensuring a safe and reliable experience for both sides of our Home Services marketplace. 

To this end, we are looking for someone with deep expertise in developing and deploying advanced machine learning (ML) models, including frontier AI and LLM models, to analyze various aspects of user and pro behavior, platform interactions and payment data to proactively address potential threats.

What you'll do 

• Initiate and drive applied science initiatives to completion with a focus on improving the trust and safety of the Thumbtack platform. This includes identifying opportunities to leverage machine learning to enhance spam and fraud detection, policy enforcement, and overall platform security. 
• Architect and deploy machine learning systems to production
• Design and execute experiments to evaluate the effectiveness of new models and strategies for improving trust and safety metrics such as reducing exposure to spam, detecting abuse and improving lead quality. 
• Analyze a wide variety of data: structured and unstructured, observational and experimental
• Collaborate with engineering, product, and operations teams to define problems, develop solutions, and ensure the successful implementation and monitoring of machine learning models for Trust & Safety.
• Maintain the right balance between speed of execution and scientific rigor when designing solutions
• Contribute to improving the overall performance of Trust & Safety models by exploring new features, algorithms, and data sources.

In order to be successful, you must bring

• Expert knowledge of machine learning techniques, including classification, anomaly detection, and natural language processing (NLP).
• Ability to effectively read, write, and debug code in programming languages such as Python
• Good knowledge of probability and statistics, including experimental design, optimization, and causal inference
• Demonstrated ability to create and drive technical and impactful roadmaps for the business, and lead seamless execution
• Ability to break down complex problems rigorously and understand the tradeoffs necessary to deliver impactful projects
• Ability to communicate clearly and effectively to cross functional partners of various technical levels

Expected salary ranges

• For candidates living in San Francisco / Bay Area, San Jose, New York City, or Seattle metros, the expected salary range for the role is currently $179,600 - $219,500. 
• For candidates living in Austin, TX or Washington DC metros or in California, Massachusetts, New Jersey, or Washington states, the expected salary range for the role is currently $161,600 - $197,600. 
• For candidates living in all other US locations, the expected salary range for this role is currently $144,200 - $186,600.

Actual offered salaries will vary and will be based on various factors, such as calibrated job level, qualifications, skills, competencies, and proficiency for the role.

#LI-Remote