Open to opportunities

Hema Sree Garikapati

@hemasreegarikapati

Message

I build production computer vision and machine learning systems for real-time inspection and predictive modeling.

United States

Message

What I'm looking for

I'm looking to build and deploy high-performance computer vision and machine learning systems, especially for real-time edge and cloud applications involving perception, robotics, and production-scale automation.

At Spirit AeroSystems, I design computer vision and machine learning models for object detection and anomaly identification in aerospace manufacturing. I improved detection accuracy for complex visual defects by 20% and reduced DNN training time by 15%.

I build full-stack ML pipelines for edge and cloud deployment, integrating real-time inspection capabilities into production environments. My work spans image segmentation, classification, depth estimation, model evaluation, and large-scale data processing.

Previously, I delivered CVML systems at Seaboard Corporation and developed predictive modeling pipelines at Ageas Federal Life Insurance, improving data processing efficiency by 25%. At Torrent Pharmaceuticals, I supported ML training, data ingestion, image analysis, and quality-control workflows.

I work with Python, PyTorch, TensorFlow, ROS, and ROS2, bringing experience in robotics, SLAM, multi-view geometry, and end-to-end ML deployment.

Experience

Work history, roles, and key accomplishments

Current

Senior Computer Vision Engineer

Current

Spirit AeroSystems

Aug 2025 - Present (1 year)

Designed and developed advanced CVML models for high-performance ML systems, focusing on perception tasks like object detection and anomaly identification in manufacturing environments. Optimized training pipelines for DNNs, achieving a 15% reduction in training time and improving model performance across large-scale datasets used for quality control.

Computer Vision Deep Learning Object Detection Anomaly Detection Cloud Systems Python PyTorch TensorFlow