Hi, my name is
Vishal Narnaware.
I build things with AI.
I’m an Atificial Intelligence engineer specializing in developing (and occasionally deploying) solutions leveraging deep learning.
About Me
Hello! My name is Vishal and I enjoy developing things that solve some of the toughest problems. My interest in deep learning started back in 2019 when I decided to explore Coursera and started Deep Learning course. Later, after years of learning I started participating in Hackathons — turns out hacking together for a societal cause taught me a lot about Deep Learning & Deployment!
Fast-forward to today, I’ve won Accenture Applied Intelligence Hackathon, Smart India Hackathon, and AI for Healthcare Hackathon. My main focus these days is building explainable, and efficient deep learning systems.
Here are a few fields and technologies I’ve been working with recently:
- Deep Learning
- Machine Learning
- Computer Vision
- Natural Language Processing
- Python
- Pytorch
- Scikit-learn
- Pandas
Where I’ve Worked
Research Intern @ Center of Excellence AIML
June - September 2022
- Evaluated linear and non-linear Machine Learning algorithms for Time Series Forecasting trained on 2500 data points
- Transformed data to stationary by Time Differencing features to remove Trend and Seasonality
- Concluded key factor was transformation of features, adding technical indicators and 1-, 2- and 3- day lags, which made project viable
- Recommended a robust and efficient pipeline for prediction, decreasing error by 10%
Some Things I’ve Built
Featured Project
interpreter.ai
A User-Friendly Translator App for Regional Languages. A webapp with 5 major features: Group Conversation, Image Translation, Walkie-Talkie Mode - Designed for quick conversation and requires only one device, Hasthamalaka - Convert sign language to regional language directly and Save & share
- Pytorch
- Django
- SpeechRecognition
- GoogleTextToSpeech
- HTML
- CSS
- JavaScript
![interpreter.ai]()
Featured Project
HateRaid
HATERAID Helps identify hate content easily. The model analyses text and image to get an idea of the content and to identify if it is hateful or not. Leverages Multimodal reasoning with Joint visual and language understanding via Parameter Sharing to understand the relation between the image and the text.
- Pytorch
- Flask
- EasyOCR
- HTML
- CSS
- JavaScript
![HateRaid]()
Featured Project
XRayd
XRayd is an AI powered web application where the user can upload an X-ray / CT scan image to get predictions from 21 lung diseases. Once the image is uploaded, it predicts disease(s) that has high probability. Healthcare being a sensitive area, AI Explainability was added to clairify predictions.
- Pytorch
- Tensorflow
- Flask
- GradCam
- HTML
- CSS
- JavaScript
![XRayd]()
Featured Project
Covid Wizard
CovidWizard is a webapp solution for giving insights to the healthcare workers.
It is a Real-time online platform that cultivates data from various sources and provide relevant information that can be sliced and diced with the help of AI as per the user’s requirements to help society fight against COVID-19
- Scikit-learn
- Azure AI
- Plotly
- Dash
- Flask
- Bootstrap
![Covid Wizard]()
Other Noteworthy Projects
view the archiveML Bootcamp
This repository contains all resources of the Machine Learning Bootcamp
Predicting Bike Sharing Patterns
Code neural network from scratch and use it to predict daily bike rental ridership.
Solve Sudoku with AI
Solve diagonal Sudoku puzzles and implement a constraint strategy called "naked twins".
Dog Breed Classifier
Given an image of a dog, the algorithm will identify an estimate of the canine’s breed. If supplied an image of a human, the code will identify the resembling dog breed.
Create your own Image Classifier
Develop code for an image classifier built with PyTorch, then convert it into a command line application.
Identify Customer Segments
Identify facets of the population that are most likely to be purchasers of AZ Direct products for a mailout campaign.
What’s Next?
Get In Touch
I’m currently looking for new opportunities, my inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!
Say Hello