Atmadeep Arya

Building new age vision based systems you say.... Count me in !!

Achievements

• Focused on developing detection and tracking models on edge devices (Nvidia Jetson Nano and AGX orin) using various yolo models and custom trackers.
• Engaged in data collection and cleaning for custom class of objects from online and in-house data sources.
• Created pipelines for ROS 1 nodes using Nvidia Jetson-inference and deepstream module for robust object detection and tracking.
• Achieved up to 3x performance boost by optimizing models and libraries for accelerated inference on server-grade hardware.
• Enabled model addition to the client’s model zoo by building testing and benchmarking suites in ONNXRuntime framework.
• Developed expertise in CNNs, NLPs, and recommender systems using onnxruntime framework

Core skill set

• Deep learning
• Computer vision
• Model optimization

Frameworks and languages

• Python
• C++

Vision for myself

• Building myself as a full stack perception engineer focused on building autonomous systems.

Background

• Project based integrated masters in computer science with 8.53 CGPA.

With over 2+ years of professional experience in field of computer vision and AI, my skill set includes model optimisation for server and edge inference using various libraries and frameworks, including but not limited to pytorch, tensorflow, ONNXRuntime using python and C++.

Masters in Computer Science with a focus on deep learning. Complemented by a project “Deep learning based methods for autonomous navigation of a drone”