abdullah muddassir

Hyderabad, India 0+ yrs exp 70 · Very Good

About

Final-year B.Tech (CSE & Data Science) graduate with hands-on experience cleaning and analysing datasets of 50K–150K records, building ML models (95–97% accuracy), and delivering dashboards in Power BI and Python. Skilled in the full data pipeline, preprocessing, EDA, modelling, and visualization, with a strong ability to communicate insights to non-technical stakeholders.

Skills & Expertise (29)

XGBoost Advanced

8.0/10

Years Exp

Power BI Advanced

8.0/10

Years Exp

Seaborn Advanced

8.0/10

Years Exp

Matplotlib Advanced

8.0/10

Years Exp

NumPy Advanced

8.0/10

Years Exp

Pandas Advanced

8.0/10

Years Exp

Random Forest Advanced

8.0/10

Years Exp

Jupyter Advanced

8.0/10

Years Exp

Google Colab Advanced

8.0/10

Years Exp

scikit-learn Advanced

8.0/10

Years Exp

VS Code Advanced

8.0/10

Years Exp

Python Advanced

8.0/10

Years Exp

GitHub Intermediate

7.0/10

Years Exp

SQL Intermediate

7.0/10

Years Exp

Git Intermediate

7.0/10

Years Exp

Tableau Intermediate

7.0/10

Years Exp

MS Excel Intermediate

7.0/10

Years Exp

Plotly Intermediate

7.0/10

Years Exp

Keras Intermediate

7.0/10

TensorFlow Intermediate

7.0/10

Dash Intermediate

6.0/10

Years Exp

CNN Intermediate

6.0/10

PyTorch Intermediate

5.5/10

NoSQL Intermediate

5.5/10

VGG16 Intermediate

5.0/10

Java Beginner

4.5/10

SAS Beginner

4.0/10

Ollama Beginner

3.0/10

LLaMA 3 Beginner

3.0/10

Work Experience

Data Science Intern

Monsac Infotech

Nov 2024 - Dec 2024

Analysed 6+ real-world datasets (UCI Heart Disease, Titanic, customer churn; 50K–150K records) to surface actionable insights used in team sprint planning. Trained Random Forest and XGBoost classification models on Titanic and churn datasets, achieving 95–97% accuracy on held-out test sets. Built 8+ interactive dashboards in Power BI and Matplotlib/Seaborn, communicating findings to both technical and non-technical audiences. Pre-processed IMDB sentiment (50K rows) and sales data with 30–40% missing values; built reusable pipeline templates that cut preprocessing time by ~50%.