-Specializing in semantic search optimization using Sentence Transformers Models with Vector Database.-Gradio application interface development, and implementing -Image classification workflows with Power Automate.

-Led inhouse products and building statistical models to extract actionable insights and drive business decisions, develop innovative solutions and drive digital transformation.- Worked on Vendor onboarding portal using API and built predictive models to forecast future trends of Inventory and sales to smoothen the supply-demand.-Monitoring model performance, evaluating model drift, and fine-tuning machine models to enhance their efficiency and present intuitive and user-friendly dashboards and reports.

-Built an app that detects the sentiment of the online classroom using live video from the webcam and real-time aggregated feedback to the instructors about the class.-Employed essential image preprocessing techniques such as image augmentation, Pixel brightness transformations, Histogram equalization, Region processing, and gray scaling to improve the image quality for better prediction.-Deployed quantized model on Streamlit and Heroku that generates sentiment of attending students and aggregates the classroom overall sentiment for the feedback.-Experimented with multiple pre-trained models such as multi-task CNN, VGG16, and Resnet50 to reduce the training time and improve accuracy, the final model built on MT CNN with a class average accuracy of 87%.-Deployed the model in the Local machine and also in the Streamlit Share app.

-Built a multi-class XGBoost classifier to predict the sentiment associated with a particular tweet and obtained the accuracy score of 74% on the test data.-Employed essential text preprocessing techniques such as text cleaning, lemmatization, named entity recognition (NER), stopword removal and tokenization.-Generated optimal document-word sparse matrix using TF-IDF Vectorizer by tuning parameters such as min_df, max_df, token pattern and n-gram range.-Deployed the sentiment model on Heroku where one can feed tweets,

-Built a clustering model using K-means to identify major customer segments on demographic and psychographic-based data to optimize the impact of marketing campaigns.-Engineered features to obtain new features such as RFM, RFMGroup, and RFMScore for getting more details about the customers' purchasing behavior.-Evaluated the optimal clusters using Silhouette score and Elbow method and leveraged the visualization library t-SNE for multidimensional scaling to visualize and validate the inter-cluster separation and intra-cluster similarities.-Checked the stability of clusters across time by generating temporal heatmaps of the customer-purchase matrix which eventually helped in auto-categorization.

- Developed a regression model using algorithms such as Ridge Regression, Random Forest, and XGBoost to predict the hourly demand of bikes and reduce the public waiting time.- Designed the data pipeline to work on, understanding the impact of features such as time of the day, weather condition, holidays and engineered important features for the model.-Tested various hypotheses, imputed missing values, encoded categorical columns, handled outliers, checked multicollinearity, and performed PCA to reduce the dimensionality.- Experimented with hyperparameter tuning techniques such as Random Search and Bayesian Optimization and achieved an R2 score of 87% using the XGBoost model and reduced the public waiting time by 10%.

-Demonstrated ability to explain complex topics in a clear and concise manner to both technical and non-technical audiences using practical use cases in automotive tech.-Extracted vehicle sensor data and Performed Exploratory data analysis for Preventive Maintenance and Troubleshooting the components based on the nominal life of parts-Experienced in the use of OBD monitors like Launch, ODIS, AUTEL, etc Did componentAnalysis for Power Train, Body Control Units and ADAS sensor data Calibration.-Worked on Preventive Maintainance and Root cause analysis for troubleshooting vehicle Performance Analysis.

Mechanical subjects like Thermodynamics, Theory of Machines, Strength of materials, Industrial and Production Engineering. Answered 100+ questions with an Overall CF score above 80%