Skip to content
View deepak2233's full-sized avatar
💭
I may be slow to respond.
💭
I may be slow to respond.
19 followers · 46 following

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: StarstruckAchievement: Pull SharkAchievement: Arctic Code Vault Contributor

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: StarstruckAchievement: Pull SharkAchievement: Arctic Code Vault Contributor

Block or report deepak2233

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
deepak2233/README.md

👱🏻‍♂️Deepak Yadav 💻

Twitter Follow Linkedin Gmail Google Scholar GitHub followers

Visitors

Hi there 👋,

As a Senior Data Scientist at Bajaj Finserv, I use deep learning to solve various natural language processing (NLP) problems. My work includes improving search results, understanding queries better, and building systems that answer questions. I am currently working on integrating large language models (LLMs) to generate better responses for FAQ chatbots, helping to resolve customer queries more effectively. I also have experience in computer vision, where I work on detecting objects and handling data drift. This involves testing and improving models to perform well with different data. I use Python, machine learning algorithms, and data science techniques to create and deploy scalable and reliable solutions.

🚀 Professional Journey:

  • Bajaj Finserv (Senior Data Scientist): At Bajaj Finserv, I'm working on a project to optimize the search functionality of Bajaj Mall's e-commerce platform using Elasticsearch, significantly improving search speed and relevance. Additionally, I worked on solutions to enhance the Bajaj Pay App by leveraging techniques such as regex for account aggregation and building an expense manager.

  • SONY India Software Center (Machine Learning Engineer): At SONY, I developed an end-to-end pose detection pipeline in C++ from Python, utilizing the Pytorch framework for bitwise serialization and ONNX for inference. This project significantly advanced the company's computer vision capabilities.

  • Stealth Start-up (Data Scientist): I researched model drift and anomaly detection for vision data using generative techniques like GAN and VAE. My work included evaluating model performance using statistical methods, developing data visualization pipelines, and performing A/B testing to assess solution effectiveness.

😎 About

  • 🌍 I'm open to a Research collaboration in the Machine Learning and Deep Learning domains.
  • 📚 I have published two research papers in NLP and computer vision under the mentorship of Prof. Baidya Nath Saha
  • 🎓 At INRIA, France, I worked on a Multimodal Emotion Detection project under the mentorship of Francoise Bremond
  • 👨🏽‍💻 I’ve completed my Research Internship in Machine Learning and Deep Learning project under the guidance of Prof. Sim Jim Yen, NDHU, Taiwan.
  • 👨🏽‍💻 Completed my Summer internship at Verzeo Pvt. Ltd. in Machine Learning Domain.
  • 🎤 Community guy who loves being involved in communities and helping students grow.
  • 🌱 I love to meet new people and Travel.
  • 👯 And Many More...

💪 Skills

C C++ HTML/CSS Python 3 Machine Learning Algorithm JavaScript (ES6) Deep Learning Data Structure and Algorithm

Hello world! 

  • 💬 Ask me about anything and everything!
  • 📫 How to reach me: message me at Whatsapp, But I may be slow to respond✍️✍️.
  • ⚡ Fun fact: I like to play Cricket 🏏, Football⚽ and eating food!
  • 😄 Pronouns: he/his/him

📈 GitHub Stats

Deepak's github stats

Pinned Loading

  1. Waste-or-Garbage-Classification-Using-Deep-Learning Waste-or-Garbage-Classification-Using-Deep-Learning Public

    This model is created using pre-trained CNN architecture (VGG16 and RESNET50) via Transfer Learning that classifies the Waste or Garbage material (class labels =7) for recycling.

    Jupyter Notebook 36 14

  2. Traffic-Signs-Recognition-using-CNN-Keras Traffic-Signs-Recognition-using-CNN-Keras Public

    There are several different types of traffic signs like speed limits, no entry, traffic signals, turn left or right, children crossing, no passing of heavy vehicles, etc. Traffic signs classificati…

    Jupyter Notebook 13 10

  3. Semantic-Segmentation-with-U-Net-for-Architectural-Data Semantic-Segmentation-with-U-Net-for-Architectural-Data Public

    Semantic Segmentation using Unet

    Python 1

  4. kNN-SVM-with-VGG16-Features-for-COVID-19-Pneumonia-Detection kNN-SVM-with-VGG16-Features-for-COVID-19-Pneumonia-Detection Public

    A scalable solution using VGG16 for feature extraction from chest X-rays and a kNN-SVM hybrid model for classification.

    Python 1

  5. GNN-Based-HIV-Molecules-Classification GNN-Based-HIV-Molecules-Classification Public

    HIV molecules classification using GNN with attention

    Python 3 1

  6. Fake_News_Detection_Using_Graph_Neural_Network Fake_News_Detection_Using_Graph_Neural_Network Public

    merge raw embedding and graph based emending for better accuracy

    Jupyter Notebook 1