I am Prudhvi Gudla

Name: Prudhvi Gudla

Profile: AI Researcher

Skills

Programming Languages and Softwares: C, Python, SQL, NumPy, Pandas, Matplotlib, PySpark, Scikit-learn, TensorFlow, PyTorch, CUDA, Flower, FastAPI, Streamlit, MLflow, LangChain, LangGraph, OpenCV, Open3D, mne, NetworkX

Libraries: ROS, Git, Github, Docker, Kubernetes, Airflow, Kafka, Linux, AWS, Arduino, ESP32

About me

I am an AI Researcher and an IIT Kharagpur graduate with experience in computer vision, robotics, IoT, and computational chemistry. At the Autonomous Ground Vehicle (Autonomy and Intelligence) Research group, I worked on embedded systems, ROS, Computer Vision, 3D Reconstruction, and SLAM. With the collaboration of a few peers in the group, we published our research on entity augmentation techniques for vertical federated learning at the GLOW@International Joint Conference on Artificial Intelligence.

As the Project Lead for Steel Eye, I developed a real-time ladle tracking system for steel plants where I worked on object detection and recognition pipelines for industry deployment, along with building a web interface integrated with an SQL chatbot for plant operators. During my research internship at Laboratoire Hubert Curien in France, funded by the prestigious Charpak Lab Scholarship, I engineered a state-of-the-art compression algorithm for JSON-LD data, improving the efficiency of constrained web applications through significant network overhead reduction.

I've also explored the intersection of AI and materials engineering, developing machine learning interatomic potentials for tungsten and gallium arsenide to enable scalable molecular dynamics simulations of radiation damage. Always eager to explore new frontiers in AI research and collaborate on projects that push technological boundaries.

Resume

Experience

Computer Vision Research Intern

Sept 2025 - Dec 2025

Akridata

  • Built a Python document parsing module leveraging metadata-based extractors, Gemini, and LandingAI agentic workflow
  • Automated the conversion of legacy inspection documents into a retrieval-optimized knowledge base for Akridata’s platform

Mechatronics and Vision Team Lead

Jul 2022 - Apr 2025

Autonomous Ground Vehicle: Autonomy & Intelligence Research Group

  • Engineered a novel entity augmentation technique for vertical federated learning, achieving a 69.5% accuracy on ResNet-18 trained on the CIFAR-10 dataset with 5% entity alignment, significantly outperforming traditional VFL approaches
  • Led the design and experimental validation of the Mixup-VFL framework. Guided a research team to implement novel label-mixing strategies that successfully utilized unaligned data, achieving up to a 72% reduction in RMSE for regression tasks with only 20% aligned data.
  • Directed an internal research initiative to deconstruct and reproduce dynamic scene 3D reconstruction methods such as HexPlane and 4D Gaussian Splatting


Charpak Lab Scholar

May 2024 - Jul 2024

Laboratoire Hubert Curien, Universite Jean Monnet

  • Have been awarded the prestigious Charapak Lab Scholarship by the French Insititute in India (IFI) for the internship
  • Worked on the Constrained Semantic Web of Things project funded by the French National Research Agency (ANR)
  • Engineered a State-of-the-art compression algorithm for JSON-LD data that provides savings of up to 95.1% in terms of network overhead
  • Developed a Python library to facilitate efficient encoding and decoding of JSON-LD data using our novel compression algorithm
  • Enhanced algorithm efficiency by integrating a Char-RNN to compress unstructured text strings using arithmetic coding

Founding Member

Jan 2024 - Apr 2025

Student Satellite Program IIT Kharagpur

  • Executed a pivotal role in the CubeSat design and overseeing sensor-related tests as a member of the Electrical Subsystem
  • Built a communication system utilizing a LoRA RFM95W with a spring antenna on the satellite and a Yagi antenna for the ground station

My Publications

Projects

Check out some of my recent projects

Achievements

Research Interests

Areas that drive my passion for innovation and discovery

Computer Vision

Robotics

Molecular Dynamics

Federated Learning

Internet of Things

BCI/HCI

My Lab

Embedded Systems Lab

My Workspace

My workspace for building embedded and IoT projects

Contact

If you're interested in brainstorming ideas, discussing research, or exploring potential collaborations, please don't hesitate to reach out to me on linkedin!