Principal Machine Learning Engineer, Account Identity

WHY SAFETY?

At Roblox, we strive to connect a billion people with optimism and civility, and the Safety organization’s mission is to become the leader in civil immersive online communities. We systematically detect, remove, and prevent problematic accounts, content and behavior, 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. Aligned and partnering with product teams, we use this toolbelt to discover new opportunities, influence and shape the product roadmap and prioritization, build safety products, and measure the impact on our community of users and developers. In doing so, we keep Roblox safe, civil, and inclusive, and we foster positive relationships between people around the world.

WHY ACCOUNT IDENTITY?

Safety and Civility is Roblox's #1 priority and our mission is to keep Roblox secure by building in-depth understanding of our users. As a principal ML engineer on the Account Identity team, you will be leading work on multi-modal age estimation (both facial and behavioral) and alternate account detection. This area is both adversarial and complex due to Roblox’s varied demographics. You'll be focused on building scalable and secure detection to ensure platform safety. You will report to the Senior Engineering Manager of the Account Identity Team.

You will:

• Lead 0-1 secure, efficient, reliable and high precision age estimation product including both facial and behavioral features


• Develop innovative detection approaches in a rapidly evolving landscape. Use a data-driven lens to think creatively, strategically, and adversarially to identify opportunities.


• Bring industry best practices and innovative research in computer vision, multi-model data fusion and big data expertise


• Up-level the ML team while also effectively collaborating with software engineers

You have:

• MS or PhD degree in Computer Vision, Deep Learning, or a related field. You have solid fundamentals in Computer Vision and classification problems.


• 8+ years of industry experience in Applied ML specifically related to facial identity/safety domain. Capable of breaking down product requirements into iterative deliverable stages. Experience in launching computer vision models into production.


• Experience with facial recognition and/or deepfake detection is a plus. This includes work on liveness detection, age estimation, anti-spoofing, and synthetic media analysis for identity verification and fraud prevention.


• Passion and Experience in anti-abuse / integrity/security domain is a plus.

You are:

• Resourceful: You're adaptable in any situation and can always find a path forward.


• Analytical: Excited to investigate large, ambiguous problems, and quantify the problems and the solutions.


• User Oriented: You always start with the needs of the user, whether they are Roblox users or internal engineers before architecting a solution.


• Team Oriented: You lift up your peers and find ways to make the team stronger. You are a sought-after mentor. You foster a culture of technical excellence.


• Mission Oriented: You stay laser-focused on the mission and are versatile, creative, and strategic to achieve it. You think long-term, and find ways to deliver short-term value on the way to get there.