Ritvik Pandey

Ritvik Pandey

Changing the world with creative thinking and code.

About Me
Hi! My name is Ritvik Pandey and I'm a
Hi! My name is Ritvik Pandey and I'm a Software Developer CFD-AI Researcher Astronomy Geek Hobby Poet

I am a firm believer of the saying that A man Never stops learning. It has always been my aim to grasp as much knowledge around me as possible, be it Engineering, Technologies, Astronomy, Politics or Bussiness Strategies and Economics. Sooner have I realised that knowledge isn't power in itself but how you use it. It has been my goal to formulate and architect solutions with greater impact in the current lines of work in the fields of Artificial Intelligence (AI), Computational Fluid Dynamics (CFD), Robotics and Software Development.

Date of birth: 17th March 1998

Phone number: (+91) 7478006888

Email: ritvik.pandey03@gmail.com

City: Bhopal | Bangalore

Freelance: Available

Resume

Languages: C++, Python, SQL, R

Development: HTML5, CSS3, JS, PHP, Android Studio

Machine Learning: Tensorflow, Keras, Pytorch, Caffe

Computational Fluid Dynamics: OpenFoam-8, Gmesh

Robotics: Arduino, RaspberryPi, Beaglebone, ROS, CAN

    Education

  • August 2016 - Present

    Indian Institute of Technology Kharagpur

    Department of Mechanical Engineering
    Dual Degree (B.Tech + M.Tech)

    CGPA: 8.24/10

    Captain: Hardware Modelling Team, Meghnad Saha Hall of Residence
    Events, Web and App Development Head: National Students Space Challenge (NSSC)
    Technology Head: Space Technology Students Society (spAts)
    Mentor: Student Welfare Group

  • August 2014 - March 2016

    Gyan Ganga International Academy

    Senior Seconday School
    Science with Maths and Physical Education

    Percentage: 88.9

    Captain: School Football Team
    Head: Astronomy Club

  • August 2003 - March 2014

    Jawaharlal Nehru School

    High School
    Language: English, Hindi, Sanskrit

    CGPA: 9.8/10

    Head: Aryabhat Astronomy Olympiad Group
    Senior member: Hindi Literary Club
    Member: Football, Hockey and Athletics teams
    Mentor: Technology Innovation Centre

    Experience

  • June 2020 - July 2020

    Honeywell India

    Systems and Control Engineer
    Bangalore, India

  • May 2019 - July 2019

    Course5 Intelligence

    Deep Learning and Vision Intern
    Bangalore, India

  • December 2017 - June 2018

    Autonomous Ground Vehicle Research Group

    Software Team Member
    IIT Kharagpur, India

  • December 2017 - June 2018

    Dewinters Opticals Inc.

    Computer Vision Intern
    New Delhi, India

My Projects

Hybrid Navigation Tech

Developed a hybrid navigation technology for indoor/outdoor use for delivery and mapping using sensor flexible SLAM implemented over AirSim simulation

Github

Pix2Flows

A Physics-informed Generative Adversarial Network (GAN) solution for approximating Reynolds Average Navier Stokes simulations around custom objects in a channel flow. This method provides a speed booast of upto 1300 times and a stoage advantage of 600 times to conventional approach

Github

3D-ShapeGAN

Tensorflow implimentation of paper "Shape Generation using Spatially Partitioned Point Clouds"

Github

DeepFoil

Computational Fluid Dynamics is a resource intensive solution to Fluid Dynamics problems while also generating enormous amounts of data. This amount of data could come handy at training Machine Learning models to either speed up or converge at approximate solutions quicker.

Github

Planogram Detection

The problem of manual counting and shelf keeping in large and small warehouses/ store-shelves is solved using Computer Vision for monitoring, localizing system and stock accounting

Github

DeepSteadyFlows

A Convolutional Neural Network (CNN) based Autoencoder model that efficiently approximates solutions for the problem of non-uniform steady laminar flows. The proposed model is able to learn complete solutions of the Navier-Stokes equations, for both velocity and pressure fields, directly from ground-truth data generated using a state-of-the-art CFD solvers

Github

DermaMentors: AI based skin prognosis

Early skin disease detection is one of the most crucial and ignored aspect in majority of developing countries. This could be solved by using medical data to train Convolution Neural Network (CNN) followed by a business plan to acquire more data and expanding the existing userbase while still keeping the startup profitable

Github

Sales Beat Plan

Any door-to-door delivery service have several contraints like maximum distance travelled, time taken, minimum revenue of the day and also has some optimization parameters like total profit, fuel prices. It is often not beneficial to cover all locations. This project aims to solve multi-route multi-vehicle optimization problem using genetic algorithm and geo-location based clustering.

Github

3D Reconstruction | Stereovision

Current stereovision tech to be as good in recreating the 3D point-cloud as any LIDAR sensor. This has also been stated by Elon Musk on being asked why Tesla card don't rely on LIDAR sensors. This projects aimed at implementing the idea of obtaining 3D point cloud of surrounding using stereovision using computer vision libraries

Github

Beer Detection

Beer bottles and can counting and detection pipeline for AbInBev Maverick Challenge. Yolov3 is used for Beer Bottle detection and Content Based Image Retrieval is used to maintain the count.

Github

hEleven

Intelligent android time scheduling app based on min-max algorithm to optimise the output given instantaneous progress and task weightage. This app is aimed for 11th hour achievers who want to maximize their productivity in least amount of time. The app gives priority wise daily schedule of target task completion and the maximum time to be alloted to each task. All in all, the aim is to ramp up the productivity even if all tasks could not be completed

Github

Baby I'm a Firefighter

PyGame based 2D firefighter game with customizable maps to to test and train preparedness towards any fire hazard in known spaces

Github

Active Learning for Object Detection

Active Learning reduces the manual labour of labelling the all of data by selectively chosing a small subset of most diverse data to be labelled relying on self evaluation metrics. This helps in achieving close to fully labelled accuracies with as less as half of total data labelled

Github

Image Stitcher

Digital Image Stitching project to empass the details of multiple zoomed image and the overview presented by their aggregate larger image for microscopic usage. Same codebase also serves the planar panaromic stitcher for camera applications

Github

Age-Gender Classifier

Deep Convolution Neural Network based age and gender classifer from facial images trained over IMDB-Wiki and indian image datasets over Caffe framework

Github

Bridge detection

Algorithm for bridge and watershed detection from SAR Satellite Imaging to design early warning system for flood and tsumani catchment area for coastal and basin areas

Github

Sperm Motality Inspection

Computer Vision based microscopic video analysis tool to inspect sperm health and motality for IVF treatment. It uses OpenCV's motion tracker module and cluster detection algorithm to rate sample on metric of motality

Github
Contact Me

Call me on

7478006888

Write to me

ritvik.pandey03@gmail.com

Send a message