About the job
About 10a Labs: 10a Labs serves as the essential safety and threat intelligence layer for cutting-edge AI laboratories, innovative AI unicorns, Fortune 10 enterprises, and prominent global technology platforms. Our expertise in adversarial red teaming, model evaluations, and intelligence collection empowers engineering, safety, and security teams to preemptively combat emerging threats and ensure the secure deployment of AI systems.
Our client's team plays a pivotal role in identifying and addressing misuse of their products, particularly new forms of abuse, while developing robust defense mechanisms. This specialized role requires additional support to monitor abuse during product launches and to architect scalable detection, review, enforcement, and risk measurement systems for identified vulnerabilities. This is a Full-Time (40 hours per week), contract position.
About The Role:
The Protection Scientist Engineer position represents an interdisciplinary opportunity combining data science, machine learning, investigation, and policy/protocol formulation. As a Protection Scientist Engineer on the client team, you will contribute to the design and construction of systems that proactively identify and mitigate abuse of AI products. This will involve establishing comprehensive abuse monitoring for new products, maintaining oversight of existing products, and prototyping defensive solutions against the highest risk threats. You will also respond to and investigate critical escalations, particularly those that existing safety systems fail to catch. This role necessitates a deep understanding of the products and data, as well as collaboration across product, policy, and engineering teams.
A strong proficiency in SQL and Python for data querying, transformation, and comprehension is essential, along with the ability to enhance prototype detection. An investigative mindset is crucial, with prior experience in identifying and addressing bad actors across various industries being a significant advantage. A background in data science, machine learning fundamentals, AI, and/or threat investigation will be beneficial.
This position is remote, based in London, UK. The role includes participation in an on-call rotation to handle urgent escalations outside of regular working hours, at least 7 consecutive days each quarter. Some detection and investigations may expose you to sensitive and distressing content, including sexual, violent, or otherwise disturbing material.
In This Role, You Will:
- Define and implement monitoring requirements for abuse during new product launches. ...

