π Data Scientist | AI Engineer | Financial Market Trader
With 2.8+ years of experience in data science, I specialize in transforming raw data into actionable insights and deploying end-to-end solutions that drive real-world impact. I'm also passionate about trading in financial markets with expertise in intraday strategies and options trading.
- π Based in Mumbai, India.
- πΌ Currently working as an Data Scientist at LTIMindtree
- π± Continuously learning new tools and techniques in AI, Machine Learning, Big Data, Cloud Platforms, and Financial Markets.
Technology : Artificial Intelligence, Machine learning, Data science, Statistics, Deep learning, NLP, GenAI(GenerativeAI), LLM (Large Language models), Predictive Modeling,RAG, Exploratory Data Analysis (EDA), Azure Cloud, Azure ML Studio, Databricks, Power BI, ADF(Azure Data Factory),AzureFacbric,MS-Excel, GitHub/Git, ETL, VectorDB
Data Manipulation and Preprocessing : Data Wrangling, Feature Engineering, Data Integration, Data Cleaning Preprocessing, Web Scraping
Client β MICROSOFT
- GPT-Based RAG Accelator :
- GPT-Based Retrieval-Augmented Generation (RAG) Accelerator
- Designed and deployed a RAG accelerator to enable conversational AI on custom datasets.
- Integrated Microsoft Azure OpenAI Services to incorporate GPT models for seamless natural language interactions.
- Implemented semantic search and data retrieval using Azure AI Search.
- Leveraged Azure Storage Containers for scalable and efficient data management, used CosmosDB for vector store.
- Developed and deployed APIs using Azure Function API.
Client: ICICI Bank
-
ICICI One Bank One Flow:
- Developed an end-to-end pipeline using Roberta NER transformers and Fasttext for intent classification.
- Achieved 90% accuracy in customer name matching using Levenshtein distance and RapidFuzz NLP.
- Created an RFM score model, helping ICICI Bank generate βΉ7 Cr revenue in 2023.
-
ICICI Mortgage:
- Built a Decision Tree model for home loan propensity scoring with LIME and SHAP for explainability.
- Applied K-Means Clustering to segment mortgage customers into Elite, Loyal, and Ace categories.
-
Other Channel Movement Prevention:
- Conducted EDA using Power BI and implemented regression techniques with TPOT.
- Predicted customers' Monthly Average Balance to assist RMs in reactivating debit card users.
-
ICICI OBOF Trade and Trade Rules:
- Developed models for identifying customers using Jaro-Winkler and Levenshtein distance.
- Utilized XGBoost via PyCaret, achieving the best classification results.
- Crypto Signals with LSTM: Bitcoin price prediction using LSTM Neural Networks.
- Web Scraping TimesJobs: Automated job listing extraction using BeautifulSoup and Selenium.
- University Shortlisting System: Admission likelihood prediction using Linear Regression.
- Fashion E-commerce Success Prediction: Predicting product success using PyCaret in an object-oriented approach.
- FaceSwapper: Fun project leveraging face-swapping techniques.
- Microsoft Certified: AI Engineer Associate Associate (AI-102).
- Microsoft Certified: Data Scientist Associate (DP-100).
- Microsoft Certified: Azure AI Fundamentals (AI-900).
- Databricks Lakehouse Fundamentals.
- Instructor-Led Training on DataBricks.
- Bachelor of Engineering in Information Technology
University of Mumbai (2018β2022) | CGPA: 8/10
- π 2nd Prize in IBM Watson Hackathon (LTI Global).
- π§ Email: [email protected]
- πΌ LinkedIn: Omkar Shirke
- π GitHub: IncognitoOmi
Financial Market Trading:
- Experienced in Nifty and Crypto Index Options Trading with a focus on intraday strategies.
- Skilled in using Open Interest (OI), OI Change, and Put-Call Ratio for market analysis.
- Proficient in Smart Money Concepts: Fair Value Gaps, Order Blocks, Liquidity Zones, and Market Structure.
If you like my work, feel free to connect and explore my repositories! π