Senior Applied AI Engineer (Multimodal Perception & Reasoning)

Lead end-to-end development of production-grade multimodal models for safety and real-time risk detection across vision, video, and language modalities.
Own the applied ML lifecycle—from data collection and labeling strategies to model fine-tuning, evaluation, deployment, monitoring, and continuous improvement in production.
Optimize for real-world conditions by balancing precision and recall, reducing false positives/negatives, and improving robustness to noise, lighting changes, occlusion, and domain shifts.
Engineer edge and cloud deployment readiness through quantization, pruning, model compression, latency and memory optimization, and hardware-aware tuning.
Build scalable training and inference pipelines with reproducibility, versioning, and reliable production deployment.
Collaborate with infrastructure and systems teams to integrate models into real-time perception pipelines and iterate based on deployment feedback.