Description
Every day, tens of millions of people come to Roblox to explore, create, play, learn, and connect with friends in 3D immersive digital experiences. At Roblox, we're building the tools and platform that empower our community to bring any experience that they can imagine to life. Our vision is to reimagine the way people come together, from anywhere in the world, and on any device.
The Safety organisation's mission is to become the leader in civil immersive online communities. We systematically detect, remove, and prevent problematic accounts, content and behaviour, and we make Roblox accounts secure and free from compromise. We cover a broad area of the tech spectrum, including machine learning, classifiers for 3D models, experimentation, automation, detection workflows, and AI-powered text filters.
As a Senior Machine Learning Engineer within the Account Identity team, you will be instrumental in developing our in-house Facial Age Estimation product and centralising age assurance controls. If you are passionate about solving complex, high-impact problems in an adversarial landscape, this is the team for you.
Responsibilities:
- Build high-precision facial age estimation and real-time liveness checks to ensure a safe user identity system.
- Develop and implement deepfake detection and 3D facial embeddings to defend against fraud and identity spoofing.
Requirements:
- Possessing or pursuing a PhD in computer science, engineering, or a related field, with a thesis aligned to Roblox’s research areas.
- Expertise in one or more areas: computer vision, facial recognition, deepfake detection, fraud detection.
- Strong coding skills with proficiency in one or more programming languages and experience with building large-scale systems.
- Experience working with large-scale data logs and tools like Spark and SQL to extract meaningful behavioural features.
- The ability to build and productionize end-to-end ML development life cycles, including data engineering and scoring algorithms.
Preferred Qualities:
- Resourceful: You're adaptable in any situation and can always find a path forward in the face of evolving threats.
- Analytical: Excited to investigate large, ambiguous datasets to find the 'signal in the noise' of billions of accounts.
- User Oriented: You understand that safety measures must be balanced with a seamless experience for our legitimate community members.
- Team Oriented: You are a sought-after mentor who lifts up your peers and builds a collaborative, high-performance environment.
- Mission Oriented: You are laser-focused on keeping Roblox safe and are willing to take bold, strategic risks to achieve a billion-user scale safely.