Sairub Naaz Email & Phone Number

Michigan, United States

- Utilized deep learning techniques to enhance the development of machine vision-based driving automation systems (autonomous driving models)- Designed and implemented a Generative AI model, using diverse datasets and integrating them into the model for optimized performance.- Built data augmentation pipelines with NVIDIA DALI, and constructed a transfer learning model to speed up the training time for testing data changes.- Implemented ETL to prepare AWS server data for annotations, and deployment of processed segmented data to the main server.- Developed an object detection model to extract focused insights from custom datasets, enhancing data analysis efficiency.- Engineered code to automate data analysis workflows, leading to increased operational efficiency, and deployed solutions to streamline processes.

Ann Arbor, Michigan, United States

- Lead the software side of the MiPACE project for Samsung to study the data from Galaxy Watches and produce metrics for Samsung’s protocol of measuring sweat data for their upcoming addition using a custom-made machine learning model in TensorFlow.- Developed an LSTM recurrent neural network model for Samsung to find the inflection threshold points using the VO2max data from their watches for increasing the precision of measuring intake of oxygen during an exercise routine.- Setup a pipeline on Apple cloud using pandas to process exercise data of Apple watch sensor kit and health kit data and produce relevant metrics at the VO2max threshold, that is inferred via the machine learning model.- Scripting FDA using PCA analysis for the GAIT data and prediction of time series data with machine learning models.- Quantification of muscle fibers using UNET model

Ann Arbor, Michigan, United States

- Automated the frame extraction process from training videos and incorporated them to train a transformer-based deep network to map the hip, knee, and ankle joints. - Developed post processing script in Python for signal processing of EMG data and mapping the processed data with transformer network output of knee flexion angle.- Integrated the workflow to connect Data Acquisition Unit (DAQ) with an amplifier and developed signal processing scripts for analyzing electromyographic signal data to map the network inferred data.

Ann Arbor, Michigan, United States

Barton Research Group Project-- Implemented the code to collect dataset for different orientations of the KUKA iiwa7 manipulator with Blender- Built an auto-encoder network to process the KUKA dataset and produce 7 belief maps, one each for every joint (key points).- Combined the resultant 2D belief maps with the known camera intrinsics of the Toyota HSR robot and forward kinematics of KUKA using OpenCV PnP to get the relative pose of the base of KUKA w.r.t HSR frame- Validated the results from the deep learning model and OpenCV with a mathematical model particularly designed to get the pose of end effector for a 7 degrees of freedom robot using AprilTags and RViz.

Noida Area, India

- Developed websites with html, css, bootstrap, js- Worked on full cycle search engine optimization for clients' websites to rank them against the competitors on search engines.- Led a team of five people to handle on-page, off-page, and technical SEO.

India

Master Of Science - Ms, Robotics, 3.8

Relevant Coursework: Mathematics for Robotics, Machine Learning, Deep Learning for Computer Vision, Robotics Systems Lab, Mobile Robotics.

Bachelor Of Technology - Btech, Mechanical Engineering, 9.4