Learn Data Skills
Sababa Saad Usmani

Sababa Saad Usmani

AI Engineer

University of Passau | Passau, Germany

Technologies

My Portfolio Highlights

My New Course

Introduction to Python

Data navigator, guiding organizations through the sea of information.

My Work

Take a look at my latest work.

course

Data Manipulation with pandas

course

Introduction to Python

course

Intermediate Python

My Certifications

These are the industry credentials that I’ve earned.

Other Certificates

Datacamp Data Analyst in Python

Datacamp Associate Data Scientist in Python

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

My Work Experience

Where I've interned and worked during my career.

University of Passau | Dec 2024 - Present

Master Thesis - LLM and Transformer-based Tender Proposal Classification

* Automated Tender Data Pipeline * Built an end-to-end workflow to scrape, parse, and preprocess tender proposals from TED Europa and other sources. * Converted unstructured PDFs into structured text, cleaned noisy data, and merged multiple datasets into a unified corpus. * LLM & NLP-based Proposal Classification * Implemented and compared traditional ML classifiers (Logistic Regression, SVM, XGBoost) with transformer-based models (BERT, Sentence Transformers). * Fine-tuned Large Language Models (LLMs) using HuggingFace Transformers for domain-specific classification. * Balanced imbalanced datasets using SMOTE/ADASYN to improve fairness and accuracy across categories. * Model Evaluation & Explainability * Conducted rigorous evaluations using precision, recall, F1-score, and classification reports. * Applied SHAP explainability to interpret model decisions and provide transparency in classification outputs. * Interactive Deployment * Developed a Gradio-based user interface for real-time tender proposal classification. * Enabled non-technical users to input tender descriptions and instantly receive automated category predictions. * Key Results * Demonstrated superior accuracy and robustness of transformer-based models compared to classical ML. * Showcased the practical feasibility of automating tender classification, reducing manual processing time. * Delivered a reproducible, end-to-end pipeline covering data scraping, preprocessing, modeling, explainability, and deployment.
Show More

Proxima AI | Sep 2018 - Oct 2021

Data Scientist

N/A

My Education

Take a look at my formal education

Master of Science in Artificial Intelligence Engineering in Artificial IntelligenceUniversity of Passau | 2026
Bachelor of Engineering in Computer Information and System in Artificial IntelligenceNED University of Engineering and Technology | 2018

About Me

Sababa Saad Usmani

Artificial Intelligence Engineer | LLM, Data Science | Python, R

Powered by

  • Work
  • Certifications
  • Courses
  • Experience
  • Education
  • About Me
  • Create Your Data Portfolio for Free