Sarab Dar

Computer Vision Engineer

HyperVisionWorld • Temps plein

Feb 2026 - Present • 8 mos

Developing and deploying end-to-end AI surveillance systems to transform passive monitoring into proactive security solutions. My work focuses on replacing manual 24/7 camera observation with intelligent, real-time anomaly detection. 𝗦𝘆𝘀𝘁𝗲𝗺 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲: Architected live video pipelines integrating IP cameras, NVRs/DVRs, and RTSP streams for seamless data ingestion. 𝗠𝗼𝗱𝗲𝗹 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁: Designing and training deep learning models for action recognition and anomaly detection, specifically targeting theft and safety protocol violations (e.g., kitchen safety compliance). 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗜𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲: Managing network configurations including port forwarding and secure remote access to ensure 24/7 system uptime and global accessibility. 𝗘𝗻𝗱-𝘁𝗼-𝗘𝗻𝗱 𝗠𝗟𝗢𝗽𝘀: Leading the full lifecycle from raw video preprocessing and data labeling to model quantization and edge deployment.

Machine Learning Engineer

MindDevLab • Temps plein

Sep 2024 - Feb 2026 • 1 yr 5 mos

- Develop and maintain backend systems in Python to support Machine learning applications. - Perform data preprocessing to Improve Model performance. - Fine-tune large language models (LLMs) for specific tasks and improved their effectiveness. - Work on agentic systems and chatbot solutions, integrating AI models into real-world use cases. - Worked with Azure Databricks to design and run automation pipelines. - Use Azure ML and AWS SageMaker for model training and deployment.

Teacher's Assistant

Under Dr. Zahid Iqbal • Temps partiel

Jan 2024 - May 2024 • 4 mos

I served as a Teaching Assistant for Dr. Zahid Iqbal in the Deep Learning course during my final semester, where I assisted students with assignments, clarified key concepts, and supported the professor in conducting tutorials and managing course activities.