Aryan Mangla

Research Engineer at NVIDIA

Aryan Mangla

I am currently working at NVIDIA. My work spans robotics, generative AI for 3D animation, and tool-calling methods. During my time here I have been co-inventor on 3 U.S. patent filings and I was awarded the NVIDIA NTech 2025 Best Paper (Global Track) in Robotics, Vision, & Autonomous Vehicles - becoming the youngest and only sole author ever to win this title at NVIDIA. I have previously worked at Intel and Astra, while being actively involved on many projects and ideas throughout my time at university.

I am quite interested in robotics & computer vision, and how intelligent systems can understand and interact with the physical world. If you'd like to discuss ideas, projects, or potential collaborations, feel free to email me or reach out on socials. I'm always open to thoughtful conversations and meaningful collaboration.

Recent News

Selected Achievements

Publications

Riemannian Geometry: Speech Detection from MEG Brain Signals towards Non-Invasive BCI

Priyanshu Vij, Aryan Mangla, Raj Mehta, Jonas De Schouwer

Poster (FR3.R2.68) · IEEE EMBS Neural Engineering (NER) 2025 · San Diego, USA · Nov 2025

Experience

NVIDIA Aug 2025 – Present
Research Engineer

Awarded NTech 2025 Best Paper (Global Track) in Robotics, Vision, & Autonomous Vehicles - youngest and only sole author ever. Ship production features in the NVIDIA App (100M+ users). Contribute to Project G-Assist. Co-inventor on 2 U.S. patent filings.

Astra Engineering - Bez Mar 2025 – Jun 2025
AI Engineer

Designed core AI architecture for solutions including data ingestion and model deployment. Fine-tuned domain-specific models achieving 55% improvement in prediction accuracy.

NVIDIA Jul 2024 – Dec 2024
Research Engineer Intern

Developed generative AI model for 2D image to animated 3D mesh conversion. Filed a U.S. patent on 3D generation and animation using GenAI, among youngest in NVIDIA to do so.

Intel May 2023 – Jul 2023
ML Research Program

Developed CNN models and deployed quantized INT8 inference with 2.1x speedup while maintaining accuracy. Leveraged Intel's TensorFlow optimizations to outperform benchmark metrics.

Education

B.Tech in Computer Science and Engineering
Manipal Institute of Technology · 2021 – 2025