Abhishek G

Sri Siddaganga PU Science College

PUC

N/A – May 31, 2021

Sri Siddaganga High School

SSLC

N/A – May 31, 2019

UBDT College of Engineering, Davanagere

Bachelor of Engineering · Electronics & Instrumentation

N/A – June 30, 2025

Hiqode Innovations

Data Analyst Intern

February 1, 2026 – June 1, 2026

India

End-to-End Data Dashboard & Sales Forecasting System

June 19, 2026 – Present

Developed a complete end-to-end data analytics solution integrating data extraction, cleaning, modeling, and visualization for business decision-making. Built a sales forecasting model using scikit-learn, applying algorithms such as Linear Regression, Random Forest, or ARIMA-style features to predict future revenue trends. Designed and automated an ETL pipeline to extract raw sales data, transform it using Python/Pandas, and load it into Power BI for reporting. Created interactive Power BI dashboards showcasing key business KPIs such as sales performance, revenue growth, customer segments, product trends, and region-wise insights. Enabled drill-down analysis in dashboards for deeper exploration of product categories, locations, and monthly trends. Implemented data refresh automation to update dashboards regularly without manual intervention. Conducted data validation and preprocessing, including handling outliers, feature engineering, and time-series formatting for accurate forecasting. Delivered actionable insights to management that supported strategic decisions on inventory planning, marketing, and resource allocation. Improved prediction accuracy through model tuning, cross-validation, and comparison of multiple ML models. Packaged the entire workflow into a scalable and repeatable pipeline for future datasets and business use cases.

AI Data Cleaning System

June 19, 2026 – Present

Designed and developed an automated AI-powered data cleaning pipeline using Python, Pandas, NumPy, and Streamlit to streamline the preprocessing workflow. Built a robust duplicate detection module using hashing, row-wise comparison, and fuzzy matching to identify exact and near-duplicate records. Integrated outlier detection algorithms such as z-score, IQR, and model-based anomaly detection to flag abnormal values in numerical features. Automated datatype validation and correction, ensuring columns follow expected formats (dates, categories, integers, floats) and converting them accurately. Added smart formatting correction rules to standardize categorical labels, remove whitespace noise, fix casing inconsistencies, and unify units. Developed an AI-driven rule engine that predicts the best cleaning action based on dataset characteristics, reducing manual decision-making. Streamlit-based UI allows users to upload raw datasets and download the cleaned version with one click. Generated a full summary report highlighting all detected issues (missing values, duplicates, anomalies, corrections applied) to improve transparency. Optimized pipeline for efficiency, achieving 70% reduction in manual cleaning time and significantly improving data quality and reliability. Enabled seamless integration with analytics workflows by exporting cleaned data in multiple formats (CSV, XLSX, JSON). Used modular, scalable architecture to allow future integration of ML-based imputations and domain-specific validation rules.

Employee Attrition Prediction & Visualization

June 19, 2026 – Present

Developed a machine learning-based attrition prediction system using Decision Tree and Logistic Regression models to identify employees at high risk of leaving. Performed extensive data preprocessing and feature engineering, including handling missing values, encoding categorical variables, scaling numerical features, and balancing class distribution. Analysed key HR metrics such as job satisfaction, work-life balance, performance rating, salary hikes, overtime, and tenure to understand attrition drivers. Trained and evaluated multiple ML models, comparing accuracy, precision, recall, and ROC-AUC to select the best-performing model. Interpreted feature importance to highlight factors that contribute most to employee turnover, enabling data-driven HR strategies. Designed an interactive Power BI HR Analytics Dashboard displaying insights such as attrition rate, department-wise attrition, age group trends, salary distribution, and prediction output. Integrated ML outputs into Power BI, enabling a combined view of prediction results and HR KPIs for actionable decision-making. Automated data refresh and report updates, ensuring HR teams always access the latest insights. Delivered clear recommendations for talent retention, employee engagement strategies, and workforce optimization based on model findings. Created a scalable and reusable pipeline to support future HR datasets and continuous attrition monitoring.