Roblox
New grad
Build facial age estimation and deepfake defenses for Roblox's billion-user safety platform
Roblox is hiring a PhD-track early-career ML engineer for its Account Identity team to build in-house Facial Age Estimation and centralized age-assurance controls, defending against deepfakes and identity spoofing using VLMs and multimodal learning. This is a research-to-production role: expect deep technical questions on computer vision/adversarial ML plus large-scale data engineering (Spark/SQL).
走进这场面试
免费 · 一场按这个岗位校准的真语音模拟
这场面试考什么
- Computer vision / multimodal learning / VLMs for facial representation and age estimation
- Deepfake detection and adversarial machine learning
- Productionizing end-to-end ML lifecycles: data engineering to scoring
- Large-scale behavioral data analysis (Spark, SQL)
- Precision/recall tradeoffs in adversarial, safety-critical systems
- Research-to-production translation from PhD thesis work
常见提问方向
Walk through your thesis or research project most relevant to facial representation, deepfake detection, or VLMs
How would you design a facial age estimation system, including how you'd evaluate precision at scale
Describe how you'd extract meaningful behavioral signal from large, noisy account-level logs
How do you think about an adversary trying to spoof or evade your detection model
Walk through productionizing an ML model end-to-end: data pipeline, training, scoring, monitoring
How do you balance catching bad actors against not degrading the experience of legitimate users
相关面试
Roblox
New grad
Senior Machine Learning Engineer (Systems), Embodied AI/NPCs, ML Platform
Roblox
Senior
Senior Product Policy Manager, Regional Policy (SEA/SA)
Roblox
New grad
Developer Engagement Representative
Airbnb
Senior
Senior Machine Learning Engineer, Customer Support Engineering
Cloudflare
Mid
Machine Learning Engineer
Coinbase
Senior