Sandeep Kadiyam Email & Phone Number

Freiburg, Bw, De

Waldkirch, Baden-Württemberg, Germany

Waldkirch, Baden-Württemberg, Germany

Developed visualizations of the inference on the Nvidia Jetson AGX Xavier device | Worked on quantizing and converting PyTorch trained models to TensorRT models | Developed a pipeline to produce 2D state maps based on the 3D annotations in a point cloud | Benchmarked object detection algorithm on Nvidia Jetson AGX Xavier and SICK Visionary-T Mini(3D Camera).

Waldkirch, Baden-Württemberg, Germany

Deep Learning Based 3D Object Detection Using Simulated ToF Data on Nvidia Jetson AGX XavierAnalyzing scalability of state-of-the-art 3D Detection models(specifically, PointPillars) for indoor logistics environments | Developed a data processing pipeline to generate depth images from various unity simulated indoor logistic environments | Installed and Integrated MMDetection3D into the detection pipeline for the Nvidia Jetson AGX Xavier | Trained the DL models on the simulated data and tested them on the real data to evaluate the generalization ability of the models.

Berlin, Germany

Calibrated two different Lidar-Camera Setups, one for a mobile delivery robot(Ouster Lidar and RGB Camera) and the other for an autonomous train(Aeye Lidar and RGB Camera) | Implemented data collection pipeline to receive and process 3D point clouds and Images from sensors | Created visualizations to identify the time synchronization issues between different perception sensors | Successfully adapted the existing clustering algorithms to meet the requirements of the detection task.

Siegen, North Rhine-Westphalia, Germany

• Developed and Deployed a deep learning model for Human Activity Recognition on bangle.js smartwatch.• Implemented Sliding window technique and Stratified splitting technique as part of the custom data loader class.• Implemented a convolution neural network for human activity recognition.• Reduced the model’s size by 96.87%, i.e., from 210.547Kb to 6.578Kb to deploy the model on the smartwatch and performed on-device inference using TensorFlow Lite.

Noida Area, India

Worked as a junior DevOps engineer | Organized the build releases and maintained continuous integration of projects using tools such as Bamboo and Git | Worked closely with the production team in deploying the final release to the production servers.

Visakhapatnam, Andhra Pradesh

Worked as a student intern in Visakhapatnam steel plant, got a deeper understanding of the working of coke ovens and coal chemical plants section.

Master Of Science - Ms, Mechatronics, Robotics, And Automation Engineering, 1.9

Bachelor Of Technology - Btech, Mechanical Engineering, 8.93

Intermediate, 95%

Ssc, 9.5