B.S.E. face recognition is a face detection and face match processes for verification/identification. Designed to high flexibility environment and proven used cases experiences. Using deep Learning engine which is developed under circumstance of hybrid use for on premise and on cloud use. Engine can connect with any standard source or non-standard source such data source from Photo, VDO clip, Streaming VDO, IP Camera or etc. The system also analyze automatically for best shot capture. The system implements image indexing, creating compact templates for searching. This in turn allows building a range of security applications such as real-time access control systems. Our system designed to perform equally well under varying lighting conditions. Recommended daylight, fluorescent and incandescent lighting. When testing on a system database, the library successfully identifies individuals in 80-99% (depand on photo enviroment).