Deva Abhishik Reddy Janga — Data Analyst · Drivetube Profile

Career Objective

Analytical and results-driven Data Analyst (MCA 2026) with hands-on experience in Python, SQL, Power BI, and Excel. Experienced in data wrangling, exploratory data analysis, statistical modeling, and dashboard creation. Delivered end-to-end projects in demand forecasting and user behavior analysis, applying machine learning models and data storytelling to generate actionable business insights. Seeking an entry-level Data Analyst role to apply technical and visualization skills to solve business problems and support data-driven decision making.

Education

Adikavi Nannaya University

Master of Computer Applications (CGPA-8.88/10.0) • Ramachandrapuram, Andhra Pradesh • Sep 2024 – June 2026

Adikavi Nannaya University

Bachelor of Computer Science (CGPA-7.77/10.0) • Ramachandrapuram, Andhra Pradesh • Sep 2021 – June 2024

Technical Skills

Programming Languages: Python,R

Frameworks and Libraries: Pandas,NumPy,Matplotlib,Seaborn,Scikit-learn

Databases: SQL,MySQL,PostgreSQL

Data and Analytics: Linear Regression,Logistic Regression,K-Means Clustering,Decision Trees,Recurrent Neural Networks,Power BI,Tableau,Microsoft Excel,Google Sheets,Data Cleaning,Exploratory Data Analysis,Feature Engineering,A,B Testing Concepts,ETL Basics,KPI Reporting,Data Storytelling

Tools and Methodologies: Jupyter Notebook,Google Colab,Git,GitHub,VS Code

Projects

Real-Time Prediction Of Taxi Demand Using Neural Networks

Tools Used: Python, Keras, TensorFlow, PyTorch, Pandas, Feature Engineering, RNN

Developed a real-time taxi demand forecasting system using recurrent neural networks, achieving R² = 0.891 (reported as 89.1% accuracy) on held-out test data.
Engineered and preprocessed large-scale taxi trip datasets spanning 263+ city zones to create time-series features, demand aggregates, and zone-level demand signatures for model training.
Built and validated a deep learning pipeline using Python with TensorFlow Keras and PyTorch, implementing sequence batching, windowing, and model checkpointing to stabilize training.
Optimized model performance through feature selection and tuning, improving generalization and reducing forecasting error during cross-validation.
Designed a low-latency inference component for live predictions, targeting and achieving ~200ms latency per prediction batch for near real-time use cases.

Spotify Listening History Analysis

Tools Used: Python, SQL, Pandas, Power BI, Data Cleaning, Exploratory Data Analysis

Cleaned and transformed 20,000+ raw streaming records using Pandas and SQL to standardize timestamps, remove duplicates, and normalize artist and track metadata for reliable analysis.
Performed exploratory data analysis to identify top artists, peak listening hours, genre preferences, and session-duration patterns, informing user engagement insights.
Built interactive Power BI dashboards visualizing listening trends, top artists, and temporal patterns to support storytelling and stakeholder review.
Improved reporting efficiency by 40% through automated data pipelines and reusable dashboard components for regular analysis and sharing.

Certifications

Data Analytics Job Simulation (Forage) — Deloitte Australia • Jul 2024

Data Analysis with Python Certification — IBM Skills Network • Apr 2026

SQL and Relational Databases Certification — IBM Skills Network • Feb 2026