PhD Student in Machine Learning and NLP. Passionate about teaching machines how to be creative.
Multi-language/multi platform fact checking powered by TigerGraph and AI. Real-time fact extraction from multiple sources into a massive knowledge graph of more than 25 million facts (and growing)!
Unlock information trapped in large collections of documents, by navigating a web app that surfaces links between documents, SDGs, people, places, organizations and themes.
Find undervalued and comparable properties using graph-based relationships powered on TigerGraph.
Reducing the noise from the search engine news search.
GraphAcademy is a personal knowledge visualization. It provides the simplest writing experience and give you a intuitive graph view of your personal knowledges.
Don't be a victim of phishing attacks!
know what you breathe !
Signver provides methods/models for signature verification - finding signatures (object detection), removing noise from signatures (image translation) and extracting semantic representations.
Parameter swapping for continual learning.
We build and train a transformer based semantic segmentation model for DL1.
We adapt RayTune for use on HPU and run distributed training with Horvod.
Using a Deep Convolutional GAN Network to Generate Synthetic Data for Medical Purposes based on Real-Life Data. For this challenge, I'm generating synthetic data for the eye disease called Cataract.
Leveraging the power of AI with a custom 15$ camera trap for wildlife conservation.
I intend to implement an ASR on EC2 Gaudi Processors
Generate short summary of long lectures.
An AI-driven stock market prediction tool based on the sentiment of the Spotify daily top songs and trained over the new powerful AWS DL1 Habana instance
Train NLP and Computer Vision models in AWS with Habana models
Cloudformation template to easily set up your dl1.24xlarge instance for training pytorch jobs.
Hecttr helps you write better issues with AI-powered assisted writing & supercharges your Jira attachments via its Image Annotation tool.
Wemotely is an all-in-one platform to create the best remote work environment for any business in only a few minutes.
An experimental aid for people with visual impairments featuring low-latency, real-time object detection using 5G Edge Computing.
AI processing application for 3D medical imaging over 5G, with visualizations & fast analytics to help medical professionals in quickly detecting cancer and diagnosing other diseases, collaboratively
Using ultra-low latency to enable the carbon removal market through AWS Wavelength and IoT services
The app shows how people can get to shelters. It is accompanied with drones and then it provides the users with information that which is the best way and real images of streets.
An AR powered application that helps lone engineers remotely collaborate in real time, provides access to an enterprise knowledgebase & enables interactive troubleshooting to circumvent sticky spots.