About Me

Master’s graduate with 2+ years of experience in data analysis, software engineering, and machine learning. Equipped with solid technical and communication skills, eager to apply analytical techniques and innovative solutions to real-world problems. Seeking opportunities to grow and contribute in dynamic tech environments.

Skills

• Programming: Python, SQL, MATLAB, HTML, CSS, Javascript, C++
• Libraries: SciKit-Learn, Tensorflow, Plotly, Pandas, Numpy, Matplotlib, Seaborn, Streamlit, Django, Flask
• Frameworks: Streamlit, Django, Flask, PyQT, PySimpleGUI
• Tools: MS Office Suite, Excel, PowerPoint, Power BI, Tableau, VS Code, PyCharm, JupyterLab, GIT, Github, Postman, Jira Software
• Platforms: Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform, Arduino, Raspberry, Firebase, Google Analytics

Education

• M.Eng, Computer & Electrical Engg. @ University Of Windsor, Ontario, Canada

  (Jan 2023 - May 2024)

• B.Tech, Electrical Engg. with Minors in Computer Science @ IITRAM, Ahmedabad

  (Sept 2018 - May 2022)

Work Experience

Developer @ Bizgull Inc (Dec 2021 - Dec 2022)

• Built full-stack web applications utilizing WordPress, Javascript, etc, with plugins, libraries, and frameworks.
• Analyzed website and web traffic data to optimize user engagement and SEO, increasing organic traffic by ~30%.
• Directed testing and debugging efforts for cloud applications, achieving a ~25% improvement in system performance.

Undergraduate Researcher (AI-ML) & Analytics @ IITRAM, Ahmedabad (Aug 2021 - May 2022)

• Developed a siren sound detection system using machine learning and computer vision with 95.7% accuracy, deployed on Raspberry Pi with a sound sensor.
• Created a custom vehicle detection and classification model using YOLO and Darknet 53 (~69% real-time accuracy).
• Designed & improved traffic light control algorithm based on a queuing theory cost function (~25% reduction in wait times).

Intern - Data Science & Analytics @ Reliance Industries Limited (May 2021 - Aug 2021)

• Performed in-depth statistical data analysis, and developed & handled 10+ PowerBI dashboards for insights.
• Formulated time series models; improved accuracy by 87% for petroleum production & close to 75% for waste segregation.
• Developed a web application for continuous monitoring, and analysis utilizing data from over 50 sensors for key stakeholders.

Web Development Intern @ Youth India Foundation (Nov 2020 - Feb 2021)

• Built an online end-to-end seamless onboarding system for new registrations during events. Handled server load-balancing during peak usage.
• Extended website’s compatibility to mobiles, and tablets by making components responsive & enabling caching. Led a team 25+ of web developers, graphic designers, and social media managers.
• Developed designs for various community websites under the organization.

Research Intern (Machine Learning) @ Indian Institute of Technology, Madras (May 2020 - Jul 2020)

• Worked on the project ‘Computer-Vision based Sketch Image Retrieval’ using advanced machine learning techniques and OpenCV library.
• Implemented an advanced version of the triplet loss function to minimize the losses in the actual and produced sketched image. Minimized the reconstruction losses up to 40% and improved the mean average precision score by approx—15%.
• Assisted with the implementation of deep sketch hashing and adversarial autoencoders on the dataset for evaluation and assessment purposes with the proposed methodologies.

Projects

1. Walmart Sales Analysis & Forecasting || View Project

Tech Stack: Python3, Scikit-Learn, Statsmodels, Plotly

• Performed extensive data analysis and visualization on data (2010–2013), combining features from over 45 stores.
• Developed robust models utilizing Random Forest and XGBoost, achieving an R2 of 0.97 for weekly sales predictions.
• Leveraged insights to develop a SARIMA-based forecasting model, achieving 89.9% accuracy in predicting future sales based on store features and regional activity.

2. Smart Traffic Control || View Project

Tech Stack: Python3, Librosa, Tensorflow, Streamlit, Arduino, SQL, YOLO

• Developed a low-cost-resource-efficient traffic control system to manage congestion and prioritize emergency vehicles.
• Designed a siren sound classification model with 95.7% accuracy, deployed on hardware with TensorFlow Lite.
• Created an emergency vehicle detection model (69% accuracy), enhanced by a queuing theory-based cost function.

3. Novel ML-Based Sleep Apnea Detection || View Project

Tech Stack: MATLAB, Python3, Scikit-Learn, MS Azure AI Platform

• Developed a novel sleep apnea detection method for nulliparous women using only airflow sensors achieving 83% accuracy.
• Leveraged Microsoft Azure to manage a 316 GB NuMOM2b database for training, validating, and testing models.
• Performed extensive data preprocessing, cleaning and analysis to develop a suitable training dataset—employed digital signal processing techniques such as wavelets, and filters for feature extraction.
• Employed digital signal processing techniques and ensemble classifiers for robust results (F1 score: 0.91).
• Implemented ensemble classifiers, data sampling, and balancing methods to ensure a robust model and facilitate result comparison and validation, achieving a testing accuracy of 83%.

4. Coursera Course Recommendation System || View Project

Tech Stack: Python3, Streamlit, Scikit-Learn, Matplotlib, Seaborn, AWS

• Implemented a course recommendation system, utilizing the 3000 courses from the Coursera Courses dataset
• Employed machine learning, natural language processing (NLP) and collaborative filtering techniques to enhance the accuracy of the recommendation system.
• Achieved a testing accuracy of 88%; further integrated the model with GUI and deployed it on the server for real-time usage.

Publications

• S. Bapodara, S. Mesvani, M. Chaturvedi, and P. Rajput, “Traffic congestion and emergency vehicle responsive traffic signal control in resource-constrained environment,” 2023 11th International Symposium on Electronic Systems Devices and Computing (ESDC), 2023. doi:10.1109/esdc56251.2023.10149873

• M. Sharma, S. Bapodara, J. Tiwari, and U. R. Acharya, “Automated sleep apnea detection in pregnant women using wavelet-based features,” Informatics in Medicine Unlocked, vol. 32, p. 101026, 2022. doi:10.1016/j.imu.2022.101026

---