About Me
AI Engineer with a Master's degree in Artificial Intelligence and Advanced Visual Computing from Ecole Polytechnique. Passionate about 3D Computer Vision, Deep Learning, and developing innovative solutions. Proven experience in multimodal learning, point cloud processing, and deploying machine learning models. Currently pursuing a PhD in collaboration with Orange and IMT Atlantique, focused on AI for 3D — developing deep learning methods for understanding, representing, and generating 3D data. Eager to contribute to challenging projects that push the boundaries of AI technology.
Work Experience
PhD Student
CurrentOrange & IMT Atlantique, France
Oct 2025 – Present
- Pursuing doctoral research in collaboration with Orange and IMT Atlantique, focused on AI for 3D.
AI Engineer (CDI)
SAMP SAS, Paris, France
Nov 2024 – Mar 2025
- Created a multimodal classifier by fusing image features and point features.
- Created synthetic images from point clouds for integrating it into the learning pipeline.
3D Computer Vision Intern
SAMP SAS, Paris, France
Apr 2024 – Sept 2024
- Developed a multimodal network for fusing point cloud features to respective image features for better segmentation of pointclouds.
- Built mapping from points to images (respective pixels) to pixel features for multimodality.
- Analysed the impact of image foundation models on point cloud segmentation.
Supervised by: Nachwa Balit, PhD & Damien Robert, PhD
Research Intern
Telecom Paris, Paris, France
May 2023 – Sept 2023
- Developed a tool to interactively edit the 3D-Skeleton of Pulmonary Airways generated from CT Scan.
- Usage of Graph Neural Networks to refine the 3D-Skeleton by using the generated segments as reference.
- Designed Geometric methods for refining connection paths in the airway.
Supervised by: Prof. Amal Dev Parakkat
Machine Learning and IOT Intern
Adihptham Technologies, Bangalore, India
Mar 2021 – June 2021
- Assisted in building the production pipeline for Computer Vision applications.
- Responsible for building edge Deep Learning and Computer Vision applications using Pytorch and ONNX.
- Worked with Arduino and Node Red for prototyping IOT products.
Education
Master of Science and Technology
Ecole Polytechnique, Paris, France
Sept 2022 – Sept 2024
Specialization: Artificial Intelligence and Advanced Visual Computing
GPA: 3.67/4
Engineering in Computer Science
Reva University, Bangalore, India
Aug 2017 – Sept 2021
Major: Computer Science and Engineering (First class with distinction)
CGPA: 8.72/10
Projects
Research Projects
Fingerprint Matching with Graph Neural Network
IDEMIA, Paris | Oct 2023 – Mar 2024
Developing a project for matching fingerprints. Leveraging Graph Neural Networks and embedded finger minutiae.
Supervised by: Laurent Kazdaghli and Victor Videau.
Fake News Detection and Correction
Nov 2020 – May 2021
Developed a novel stance detection model for news articles and web-scraped data for authenticity. Awarded Best Paper at Reva University, 2021.
Course Projects
Evolution in Wonderland - Unity3D Game
Sept – Oct 2023 | Course: Advanced 3D Graphics
Developed a game using terrain brushes, Inverse Kinematics, genetic algorithms, and neural networks for animal brains.
Figure from Cheng et al., "Per-Pixel Classification is Not All You Need for Semantic Segmentation", NeurIPS 2021
MaskFormer Implementation
Oct – Dec 2023 | Course: Deep Learning
Studied, analysed and Implemented the MaskFormer paper from scratch for semantic segmentation.
Figure from Radford et al., "Learning Transferable Visual Models From Natural Language Supervision" (CLIP), 2021
Bird Species Retrieval (CLIP & Blip-2)
Feb – Mar 2024 | Course: Computer Vision
Implemented CLIP and Blip-2 for bird species retrieval from CUB-200-2011 dataset, using linear probing.
Figure from Wiersma et al., "DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds", SIGGRAPH 2022
DeltaConv Analysis
Nov – Dec 2023 | Course: Analysis and Deep Learning on Geometric Shapes
Analysed and implemented DeltaConv on Point Clouds, studying differences with KPConv.
Point-based radiance field render, after Zhang et al., "Differentiable Point-Based Radiance Fields for Efficient View Synthesis", SIGGRAPH Asia 2022
Differentiable Point-Based Radiance Fields
Jan – Mar 2023 | Course: Image Synthesis
Analysed and implemented the paper for efficient view synthesis from scratch.
DDPG learning curve, after Lillicrap et al., "Continuous Control with Deep Reinforcement Learning", 2015
Continuous Control with DRL
Oct – Dec 2023 | Course: Deep Reinforcement Learning
Analysed DDPG algorithm for physical games using visual data, studying differences with DQNs.
Technical Skills
Languages
Developer Tools
Libraries/Frameworks
Extras
Patent
Air Pollution Control System Using Tunnels
Jan 2021 | Indian Patent Office (No: 202041023584)
Novel design for metropolitan cities, simulated using CFD and optimized with DNNs.
Achievements
Hack Zurich Winner
Sept 2022
Won "Best Project for Huawei-Sunrise-Swissski" for creating a monitoring tool for skiing athletes.
Relevant Coursework
- Machine Learning & Deep Learning
- Deep Learning
- Deep Reinforcement Learning
- Analysis and Deep Learning on Geometric shapes
- Advanced ML & Autonomous Agents
- Image Analysis & Computer Vision
- Advanced 3D Graphics
- Image Synthesis
- Soft Robotics, fabrication and Control
- Navigation for Autonomous Systems
- Computer Animation
- Natural Language Processing
- Real-Time AI in Video Games
- Intelligent Agents
- Advanced Data Structures & Algorithms