🚀 Results-driven Data Scientist with 4 years of relevant experience, specializing in Generative AI, Azure ML, and ML solution design. AWS Verified Machine Learning Specialist, with expertise in developing scalable AI solutions across supply chain management, Text2SQL agents, and healthcare domains. Proficient in deploying enterprise-level ML applications on Azure, driving operational efficiency and business impact.
Skilled in leveraging cutting-edge AI technologies in Machine Learning, Deep Learning, and NLP, with a strong focus on RAG Architecture, Prompt Engineering, and Information Retrieval. Proven ability to collaborate with cross-functional teams, optimize AI workflows, and drive revenue growth. Passionate about AI solution architecture and LLM fine-tuning, enabling organizations to innovate and scale with AI-driven strategies.
⸻
⚡ 4.5+ years of expertise in ML, NLP, Deep Learning, and Python.
⚡ Specialized in NLP, BERT, Sentiment Analysis, Topic Modeling, RAG Architecture, and Prompt Engineering.
⚡ Expert in Prompting Techniques (CoT, ToT) for NP-hard problems, including RAG using LangChain, VectorDB, Embeddings, and Information Retrieval from unstructured documents.
⚡ Strong foundation in Statistics, Classical ML (Decision Trees, XGBoost, GBM), Optimization, and Deep Learning.
⚡ Experienced in Azure Infrastructure, designing and deploying highly scalable systems in Microservices architecture (Service Bus, Azure Functions).
⚡ Leading a team of 2 on 4 GenAI PoCs, including designing AI-enabled multi-agent Tax Copilot Chatbot Systems.
⚡ Skilled in Open Source LLM fine-tuning (PEFT, LoRA) and Transformer architectures.
⚡ Top 5% Rank holder in ML/DL Hackathons (HackerEarth).
⚡ Published 2 h-index Conference Papers.
⚡ Skilled in Azure Document Intelligence.
⸻
• Languages: Python, SQL, R, LaTeX
• Frameworks: Scikit-learn, LangChain, VectorDB, NLTK, SpaCy,
• Keras, Flask:, Streamlit, R-Shiny, RAG
• Cloud & Tools: AWS ML, GIT, Huggingface, Bitbucket, MySQL, Databricks, Azure ML Studio, Azure Document Intelligence
• Platforms: Web, Windows, Azure
• Soft Skills: Leadership, Event Management, Public Speaking, Time Management\
Resume: Resume Link
- Generative AI: Proficient in Large Language Models (LLMs) such as GPT and RAG, capable of generating human-like text and performing retrieval-augmented generation tasks.
- Natural Language Processing (NLP): Skilled in NLP concepts including embeddings, information retrieval, and sentiment analysis.
- Machine Learning Operations (MLOps): Experienced in deploying ML models using Azure ML and managing end-to-end deployment pipelines.
- Time Series Forecasting: Building dynamic pipelines for time series forecasting using state-of-the-art techniques.
- Cloud Deployments: Well-versed in deploying AI solutions on the Azure cloud platform, ensuring scalability, reliability, and security.
Feel free to reach out if you want to discuss AI, ML, or anything tech-related!
I have publised two research papers. You can see the edescription below.
| No. | Description | Published | GitHub Repo | Link to Publication |
|---|---|---|---|---|
| 1. | Detection of Coronavirus (Covid19) using Deep Convolutional Neural Networks with Transfer Learning using chest X-Ray Images | Under Review | Code | Link |
| 2. | MLAI: An Integrated Automated Software Platform to Solve Machine Learning Problems | Published | Link | Link |
| 3. | Topic Classification using BERT: A Deep Dive into Natural Language Processing | Published | N/A | Link |
| 4. | Vision Transformer (ViT) Decoded: A New Era in Computer Vision | Published | N/A | Link |
ML Competitions on HackerEarth
The following table contains all the code bases of the competitions that I participated on HackerEarth. The original repository is uploaded in my GitHub account in their respective repos. However, for easier browse through specific problem and solutions, this table may come handy to you. Keep Learning!
| No. | Challenge Name | GitHub Repo | Type | Position | LeaderBoard |
|---|---|---|---|---|---|
| 1. | HackerEarth Machine Learning challenge: Calculate the severity of an airplane accident | Solution* | Classification | 1 st / 7449 teams (Winner) | Link |
| 2. | HackerEarth Deep Learning Challenge: Snakes in the hood | Solution* | Classification | 13 th / 3389 teams (top 0.3 %) | Link |
| 3. | HackerEarth Machine Learning Challenge: Carnival Wars | Solution* | Regression | 23 rd / 2144 teams (top 1%) | Link |
| No. | GitHub repo | Description | Category |
|---|---|---|---|
| 1. | Text2Sql-Agent | The Text2SQL Chatbot is an advanced tool that allows users to interact with a SQL-driven assistant to query and explore data seamlessly. The chatbot can generate SQL queries from natural language input, execute them, and provide tabular responses with insights. Additionally, it has the capability to generate charts based on user queries. | GenAI Text2SQL Agent |
| 2. | Self Driving Car using Raspberry Pi & Arduino | We optimized Canny Edge Detection algorithm incorporating with pixel intensity distribution} to identify parallel pathways which is better than existing technology.The car can detect Road Signs and can also interpret traffic Light Signal using OpenCV and controls itself accordingly by forwarding impulses to Arduino to control rotor's rotational speed | Lane & Object Detection |
| 3. | Alpha AI : Automated ML Web App | Alpha AI is an automated web platform that helps students & data scientists to dig deeper into the data, finding insights and comparing different ML models based on two use cases(Regression & Classification). | Auto ML |
| 4. | Movie-Recommendation-System | This is a Streamlit based Movie Recommendation System build on Content Based Filtering & Demographic Filtering (Plot Description , Popularity/ Genres/ Cast/ Director). Used TF-IDF Vectorizer . | Recommendation System |
| 5. | AutoScan - Text Extraction Application | Working on a Text Extraction & Labeling Application which could extract and label texts From Business Cards, PDFs and Images using tesseract, and BIO Tagging. Here is the outcome of the code. . | OCR |
| 6. | Table Transformer : Detection Of Tables Uisng Table Transformer | This Repository contains code for Detecting Tables from Images, Cropping The Tables and perform Table Extarction. | OCR |
📈 My GitHub Stats