Lead AI & Computer Vision Engineer
Evercam Construction Cameras
2019 – Present
Lead the design and development of large-scale real-time video analytics systems for construction site monitoring across distributed camera networks.
- Designed and deployed real-time computer vision models for object detection, tracking, and behavioral analysis on live camera feeds.
- Built end-to-end ML pipelines covering dataset generation, annotation workflows, training, evaluation, and automated deployment.
- Developed video analytics features including crane activity detection, material classification, pose estimation, human blurring, and movement detection.
- Optimized deep learning models for edge deployment on NVIDIA Jetson devices, reducing inference latency and enabling real-time processing.
- Architected scalable backend services and APIs using Python and FastAPI to support large-scale video analytics platforms.
- Designed distributed data processing pipelines and messaging systems using RabbitMQ and PostgreSQL/TimescaleDB.
- Implemented monitoring and observability tools using Grafana, Loki, and Prometheus for production ML systems.
- Delivered multiple product features including Smart Search, PPE detection, Gate Reports, and AI Copilot capabilities.
- Improved system robustness through distributed processing, queue-based architectures, and fault-tolerant data pipelines.
- Mentored junior engineers and contributed to architecture decisions for production AI infrastructure.