|
Pranay Lendave
Hello, I'm Pranay Lendave. I am currently working as an Analyst at Deloitte. I obtained my bachelor's degree in Electronics Engineering from KJ
Somaiya College of Engineering under the guidance of Prof. Sushma Kadge. In my final, I had the
opportunity to participate in a semester-long internship at IIIT, Sricity, where I collaborated
closely with Prof. Pavan Kumar BN. I worked closely with Prof. Palash Ingle.
My research interests lie in 2D and 3D computer vision, multi-model learning, neural rendering and
LLM agents.
Email /
CV /
Google
Scholar /
GitHub
|
|
Publications
|
MT3DNet: Multi-Task Learning Network for 3D Surgical Scene
Reconstruction
|
|
Mithun Parab, Pranay Lendave, Jinman Kim, T.Q.D. Nguyen, Palash Ingle
|
| arXiv preprint (arXiv:2412.03928) |
|
Paper
|
|
|
A Comprehensive Study on LLM Agent Challenges
|
|
Palash Ingle, Mithun Parab, Pranay Lendave, Pavan Kumar BN
|
| AAAI 2024 Spring Symposium (Stanford, USA) |
|
Paper
|
|
|
|
|
A Novel Approach to Weed Detection Using Segmentation and Image Processing
Techniques
|
|
Shafina Charania, Pranay Lendave, Janak Borwankar, Sushma Kadge
|
| WCONF 2023 (Raipur, India) |
|
Paper
|
|
|
|
|
Experience
|
Business Technology Analyst at Deloitte USI
Oct. 2023 – Present
|
• Lead QA overseeing the development of autonomous customer query resolution systems,
• Including voice bots and chatbots, leveraging AWS, Genesys, and Salesforce solutions.
|
|
|
Undergraduate Researcher at IIIT, Sricity
Jan. 2023 - Jun. 2023
|
|
Advised by Prof. Pavan Kumar BN
|
• Worked as an AI researcher on 3D object detection using deep learning techniques
• Evaluated state-of-the-art models across benchmark datasets such as KITTI, WAYMO, and nuScenes.
|
|
|
Projects
|
|
Text to Video Synopsis
|
|
Developed an advanced video analysis tool that uses text prompts as input. The system consists of state-of-the-art computer vision models (OWL-ViT/Florence 2 for detection, SAM for segmentation) and video synopsis algorithms for surveillance, content analysis, and object tracking.
|
|
Github
|
|
|
|
GPS Tracker and SOS Notifier
|
|
Developed an IoT-based system using a WiFi-enabled microcontroller with sensors (MPU6050, NEO-M8N) for tilt, motion, and positioning monitoring. Integrated ThingESP and Twilio platforms for real-time communication with WhatsApp users, enabling emergency notifications and data display on an OLED screen.
|
|
Github
|
|
|
|
Smart Conveyor Belt for Honey Jar Detection and Sorting
|
|
Designed a smart conveyor belt system utilizing YOLOv5 and ESP32 microcontrollers for wireless detection and sorting of honey jars based on condition, connected via a local Wi-Fi network.
|
|
Github
|
|
|
|
Car Segmentation and adding Virtual Background using U2NET
|
|
This project utilizes the U2Net deep learning model to perform car segmentation in images. After segmenting the car from the background, it allows you to replace the background with a virtual background of your choice.
|
|
Github
|
|
|
Volunteer roles and responsibilities
|
|
|
Technical Head
|
|
Electronics Engineering Students Association, KJSCE
|
|
Organized workshops and seminars on cutting-edge technologies, collaborating
with industry experts to provide guidance.
|
| Webpage
|
|
|
|
|
Electronics Head
|
|
The Marine Robotics Team (TMRT), KJSCE
|
|
Led a team of four members to build a navigation system for an autonomous underwater vehicle.
|
| Webpage
|
|
|
|