Shiliveri Sai Jyothi
Entry-level Data Analyst • Bangalore, India • s*************@gmail.com • +91*******792 • linkedin.com/•••••
Career Objective
Recent B.Tech graduate in Computer Science with hands-on project experience in data analysis, NLP-based classification, and computer-vision OCR pipelines. Proficient in SQL and Python (Pandas, NumPy) and experienced with Power BI and Excel for reporting and visualization. Built and validated classification and real-time detection systems that improved content accuracy and reduced engagement with harmful content in project settings. Seeking an entry-level Data Analyst role to translate data into actionable insights, optimize reporting workflows, and support data-driven decision-making.
Education
Vaagdevi Institute of Technology and Science
Bachelor of Technology in Computer Science & Engineering • January 2020 – May 2024
Sri Shiridi Sai Junior College (Board of Intermediate Education)
Intermediate • June 2018 – May 2020
Technical Skills
Programming Languages: Python
Frameworks and Libraries: Pandas,NumPy,scikit-learn,OpenCV,Matplotlib,Seaborn
Databases: SQL,SQL queries,Relational data concepts
Data and Analytics: Data cleaning,Data preprocessing,ETL concepts,Supervised classification,Model evaluation,Power BI,Excel
Tools and Methodologies: Jupyter Notebooks,Statistical analysis,Hypothesis testing
Skills: Text preprocessing,NLTK,spaCy
Computer Vision & OCR: Tesseract OCR,Image preprocessing
Projects
Social Media Misinformation Detection and Democratic Verification Mechanism Design | January 2024 – May 2024
Tools Used: Python, Pandas, NumPy, NLTK, spaCy, scikit-learn, Power BI, Statistical analysis
- Collected and cleaned large-scale social media interaction datasets to prepare features for classification and trend analysis, reducing noisy entries by 35%.
- Engineered textual features and trained supervised classification models using scikit-learn to detect misleading content, achieving high precision in test validations.
- Designed a democratic verification mechanism that cross-referenced multiple sources and consensus rules, improving verified content accuracy by 28%.
- Developed NLP pipelines (tokenization, stopword removal, lemmatization) and evaluated models with precision/recall metrics to prioritize low false-positive rates.
- Built dashboards in Power BI to visualize misinformation trends and statistical impact on public opinion, enabling clear stakeholder reporting.
Automation Recognition of License Plates | June 2023 – September 2023
Tools Used: Python, OpenCV, Tesseract OCR, Image preprocessing, Pandas
- Implemented an automated license plate recognition pipeline using OpenCV for detection and Tesseract for OCR to extract plate characters reliably.
- Applied image preprocessing techniques including adaptive thresholding and edge detection, improving detection throughput and robustness under varied lighting.
- Optimized character segmentation and OCR integration to increase recognition speed and real-time processing capacity by ~30%.
- Validated system performance with cross-validation on diverse datasets and quantified robustness improvements (~15%) in challenging lighting conditions.
Certifications
Mathematics for Data Science — Simplilearn
SQL for Data Analysis — Simplilearn
Powered by Drivetube · Create your own profile at drivetube.ai