Ankit Kumar

DATA ANALYST

[email protected] +91 8105655092

Resume

Hello, I am

Ankit Kumar.

2

PowerBI
Projects

About Me

I am Ankit Kumar, a Data Analyst enthusiast with a background in Electrical and Electronics Engineering from Bangalore Institute of Technology. My schooling journey began at Navy Children School, Visakhapatnam. Over time, I’ve nurtured a passion for data-driven insights and problem-solving through hands-on experience in machine learning and data visualization projects. I've built models for credit risk prediction, house price forecasting, and business performance analysis. Additionally, I’ve developed interactive Power BI dashboards for supply chain management, HR analytics, and sales insights. My current internship at YBI Foundation allows me to apply and expand my analytical skills, contributing to impactful business decisions. I'm constantly learning and seeking opportunities to bridge data and actionable outcomes through thoughtful analysis and visualization techniques.

Key Skills

Power BI

Tableau

Excel

pandas

scikit learn

matplotlib

Seaborn

Feature Engineering

Data Preprocessing

Hyperparameter Tuning

Model Optimization

Data analysis

Report Building

Strong Stakeholder Management

People Management

Communication

Leadership

MySQL

Fabric

Jira

My Projects

HEALTHCARE PREMIUM ESTIMATION USING ML

Domain/Function: HEALTH SECTOR

CREDIT RISK MODELLING USING ML

Domain/Function: FINANCE SECTOR

BUSINESS ANALYTICS

Domain/Function: BUSINESS ANALYSIS

HR ANALYTICS

Domain/Function: HR ANALYTICS

My Experience

WORK EXPERIENCE:

Data Analyst Intern @ YBI Foundation January | 2025 – Present

• Extracted, cleaned and analyzed datasets comprising "50,000+ records" to deliver data-driven insights enhancing operational efficiency by 25%.

• Developed interactive dashboards and visualizations using Power BI to illustrate key data trends, improving decision-making processes.

• Streamlined data preparation workflows, reducing processing time by 20%.

Academic Mentor @ Pigeon EdTech August | 2021 – June | 2022

• Optimized Excel to analyze data for 500+ students, generating performance reports that improved academic tracking efficiency by 15%.

• Discovered key performance trends and suggested actionable interventions to enhance student outcomes by 25%.

• Streamlined data collection processes, reducing errors by 10% and ensuring data consistency across multiple datasets.

• Collaborated with academic teams to implement data-driven strategies for personalized learning plans.

Awards & Certificate

Excel: Mother of Business Intelligence

Get Job Ready: Power BI Data Analytics for All Levels 3.0

AI for Everyone

Python: Beginner to Advanced For Data Professionals

Master Machine Learning for Data Science & AI: Beginner to Advanced

SQL Beginner to Advanced For Data Professionals

Let's Connect

Feel free to get in touch with me. I am always open to discussing new projects, creative ideas or opportunities to be part of your visions.

Mail Me

[email protected]

https://www.linkedin.com/in/kmrak

Call Me

+91 8105655092

GitHub

https://github.com/kmrak47

HEALTHCARE PREMIUM ESTIMATION USING ML

Domain/Function: HEALTH SECTOR

Healthcare Premium Estimation Using Machine Learning involves building predictive models to estimate the cost of health insurance premiums based on various factors such as age, gender, health conditions, lifestyle choices, and more. By leveraging historical data, machine learning algorithms like reg

View Project
Engagement

View Project
on GitHUB

View Project
on

Health insurance covers medical expenses by reducing out-of-pocket costs for treatments, offering financial security and access to necessary care.

Developed a machine learning model to predict health insurance premiums, optimizing accuracy based on customer demographics and policy details.

Project Details

• Objective: Predicted health insurance premiums with 99.4% accuracy, considering factors like age, smoking habits, BMI etc.
• Data Processing: Established a robust system for cleaning and normalizing over 50,000 records, enabling more reliable dataset outputs; the resulting data architecture now supports advanced analytical models utilized by data scientists in ongoing projects.
• Algorithms: Employed linear regression, XGboost with 98% accuracy, and Ridge regression.
Evaluation: Employed GridSearchCV and RandomizedSearchCV, achieving 99.7% accuracy through hyperparameter tuning.
• Outcome: Delivered a predictive model to assist insurance companies with a accuracy of 99.7% estimating premium costs, enhancing decision-making processes.

CREDIT RISK MODELLING USING ML

Domain/Function: FINANCE SECTOR

Initiate a model with 96% accuracy to assess loan approval risk based on factors like age, employment status, income, residence, CIBIL score, and loan amount.

View Project
Engagement

View Project
on GitHUB

View Project
on

Credit risk predicts the likelihood of a borrower defaulting on a loan, aiding lenders in making informed decisions and minimizing potential financial

Built a machine learning model for credit risk prediction, assessing borrower profiles to predict default probabilities and improve lending decisions.

Project Details

• Objective: Initiate a model with 96% accuracy to assess loan approval risk based on factors like age, employment status, income, residence, CIBIL score, and loan amount.
• Data Processing: Analyzed and processed 50,000+ loan applications, leading to a 96% improvement in risk prediction accuracy.
• Algorithms: Applied logistic regression, random forests, and XGB, reducing loan default prediction errors.
Evaluation: Employed KS Statistic, Gini Coefficient and Optuna achieving 99% accuracy through fine-tuning of hyperparameters
• Outcome: Formulated a risk assessment tool that enabled financial institutions to make informed loan approval decisions, resulting in a 20% reduction in default rates

PowerBI Project

BUSINESS ANALYTICS

Domain/Function: BUSINESS ANALYSIS

Finance View

Sales view

Project Details

Problem Statement:
• Atliq Hardware, a global manufacturing unit, faced significant losses.
• The task involved analyzing the profit and loss (P&L) statement and evaluating various key performance metrics to identify underlying causes and
recommend actionable strategies.
Finance Analysis:
• Evaluated that the desktop segment dominated the market by 151%, with the notebook segment following at 82%.
• In 2022, gross sales surpassed the previous year by 5.71 billion, with net sales increasing by 2.91 billion.
Sales Analysis:
• Estimated net sales at 4.9 billion with a gross margin of 1.8 billion, representing a gross margin percentage of 37.63%.
Market Analysis:
• Identified that Asian markets outpaced North American markets by 0.72%, with Asian regions emerging as the largest market.
Supply Chain Data Analysis:
• Evaluated net errors of 2 million and reducing overall absolute errors by 25 million. Attained a forecast accuracy rate of 79.83%.

Live Interactive Dashboard

PowerBI Project

HR ANALYTICS

Domain/Function: HR ANALYTICS

I developed an HR Analytics Power BI dashboard for Atliq Technologies, featuring key KPIs like WFH%, SL%, and Presence% by weekday. The project included dynamic tables and visualizations for real-time insights, ensuring seamless updates without manual intervention, enhancing decision-making.

view

HUMAN RESOURCE

Project Details

I developed an HR Analytics Power BI dashboard for Atliq Technologies, focusing on key employee performance metrics. The dashboard featured KPIs such as Work From Home Percentage (WFH%), Sick Leave Percentage (SL%), and Presence% across weekdays, visualized through tables and line charts for trend analysis. A dynamic template ensured seamless data updates without requiring manual transformation, making the solution scalable and efficient for ongoing use. The dashboard also included an Employee Details Matrix, providing granular insights into workforce patterns. This project addressed a critical challenge by streamlining HR reporting and enhancing the organization’s ability to make data-driven decisions regarding employee productivity and attendance trends. The comprehensive visualizations empowered stakeholders to identify operational patterns and optimize workforce management strategies effectively.