Content Moderation Technical Lead Manager

About the Role
You'll lead the design, training, and deployment of computer vision models that classify user-generated images and video at scale. You'll build the team, choose the architecture, and own the platform from research to production.

Key Responsibilities

• Define the technical architecture and roadmap for an in-house image and video moderation platform.
• Build, train, and fine-tune computer vision models to detect categories such as nudity, sexual content, violence, gore, weapons, drugs, hate symbols, and self-harm.
• Develop video moderation pipelines, including frame sampling, scene detection, and temporal analysis.
• Build detection for AI-generated and manipulated media, such as deepfakes.
• Design data collection, labeling, and quality-control processes, and manage labeling teams or vendors.
• Deploy models to production with low latency and high throughput, using GPU inference, model optimization, and autoscaling.
• Build APIs that return category scores and confidence levels to our apps.
• Implement hash matching for known harmful content, and integrate specialized child-safety tools such as Thorn Safer or PhotoDNA.
• Support US compliance needs, including NCMEC reporting workflows and TAKE IT DOWN Act removal and duplicate detection.
• Measure model performance through precision, recall, false positive rates, and bias testing, and improve it continuously.
• Hire, lead, and mentor ML engineers, data engineers, and backend engineers.
• Required Qualifications
• Master's or PhD in Computer Science, Machine Learning, or a related field, or equivalent industry experience.
• 7+ years of experience in software or ML engineering, including 3+ years building computer vision models in production.
• 2+ years leading technical teams or projects.
• Deep expertise in PyTorch or TensorFlow and modern vision architectures such as CNNs, Vision Transformers, and CLIP-style models.
• Experience with transfer learning and fine-tuning pretrained models.
• Experience deploying models at scale, using tools such as Triton, TorchServe, ONNX, TensorRT, Docker, and Kubernetes.
• Strong Python skills and solid backend engineering fundamentals.
• Experience with cloud GPU infrastructure on AWS, Google Cloud, or Azure.
• Experience building data pipelines and managing large labeled datasets.

Preferred Qualifications

• Previous work on content moderation, trust and safety, or integrity at a social media or user-generated content company.
• Experience with video understanding models and real-time or live-stream processing.
• Experience with perceptual hashing (such as PDQ or pHash) and near-duplicate detection.
• Experience with multimodal models that combine image, video, and text.
• Knowledge of deepfake and AI-generated content detection.
• Familiarity with US online safety requirements (NCMEC reporting, TAKE IT DOWN Act, COPPA).
• Published research or open-source contributions in computer vision.

Pay: $400.00 - $500.00 per month

Work Location: Remote