Machine Learning Engineer
Core
Design and build machine learning systems for real-time moderation of live video interactions, detecting unsafe, inappropriate, abusive, or policy-violating content.
Role type
Senior IC machine learning engineer (computer vision & multimodal)
Builds
Scalable inference infrastructure and moderation architectures for live video platforms
Domain
Trust & safety / Content moderation / Computer Vision
Deliverable
production ML models
Required skills
Python, PyTorch, computer vision (classification, detection, OCR, video analysis), GPU/CUDA optimization, Docker, Linux, cloud infrastructure, software engineering fundamentals (reliability, observability, testing)
Preferred skills
trust and safety, content moderation, fraud detection, live/near-real-time video processing, OpenCV, FFmpeg, WebRTC, Hugging Face Transformers, multimodal models, ONNX Runtime, TensorRT, Triton, model quantization, PaddleOCR, EasyOCR, Tesseract, MLOps, adversarial ML detection, shadow deployments, A/B testing, Redis, Elixir
Technologies
PyTorch, CUDA, Docker, Linux, Redis, OpenCV, FFmpeg, WebRTC, ONNX Runtime, TensorRT, Triton, Hugging Face Transformers
Responsibilities
Develop computer vision and multimodal models for content detection; optimize precision, recall, latency, throughput, and cost; build scalable inference infrastructure; design moderation architectures with confidence thresholds and escalation paths; investigate adversarial behavior and bypass attempts; deploy and monitor models in production; create tools for model decision review and policy improvement; maintain documentation for architecture, datasets, and experiments.
Seniority
Senior, hands-on IC