• Enhanced a full-stack platform connecting service seekers with professionals by refining UI/UX, expanding search
functionality, and streamlining payment workflows across multiple user scenarios.
• Designed and developed React.js components while optimizing Node.js/PostgreSQL APIs, integrating Google &
Facebook authentication to strengthen login security and ensure seamless cross-platform access.
• Collaborated with a 5-member Agile team on code reviews, debugging, and deployment to ensure application stability.

• Debugged code across 5+ live projects, improving system reliability and reducing reported issues by 20%.
• Led project management initiatives, planning, organization, and scheduling to ensure timely project completion.
• Guided a group, offering technical direction and support to enhance team performance and project outcomes

∗ Mentor: Supported 5-10 teenagers in both academic and personal development, offering guidance, encouragement, and
consistent check-ins to help them succeed.
∗ Supervisor: Promoted to oversee the mentorship program, coordinating 10 mentors, monitoring +30 mentee progress, and
ensuring a supportive environment for participants.

∗ Built a machine learning pipeline to classify 1,000+ text reviews of musical instruments into positive/negative sentiment.
∗ Applied NLP preprocessing (tokenization, stemming, lemmatization) and trained a Random Forest model.
∗ Achieved 76.5% accuracy and 80% F1-score, demonstrating balanced precision and recall in text classification.

∗ Implemented a clustering model to segment retail customers into 4 groups, achieving a Silhouette Score of 0.54.
∗ Conducted comprehensive data preprocessing, including feature engineering and scaling, to enhance model accuracy.
∗ Leveraged Python, Scikit-learn, and Seaborn for analysis and visualization, revealing insights for customer behavior.

Built and trained a feedforward neural network on the MNIST dataset, achieving 97% classification accuracy across 10 handwritten digit classes.

Tuned hyperparameters (epochs, learning rate, batch size) and visualized model performance, demonstrating strong understanding of deep learning fundamentals.

https://github.com/mehmet-akif/The-Training-Hub

∗ Developed a dynamic sports training platform with 6 sports and 3 registration packages each.
∗ Created custom graphics and ensured responsive design for seamless experience across devices.
∗ Integrated MongoDB databases with Node.js and Express.js, using REST API, deployed on the university server.

∗ Developed a comprehensive bug report system as part of a team using React for the frontend, Node.js for the backend,
and Supabase for the database.
∗ Implemented features to insert, retrieve, and edit bug reports, ensuring smooth data flow and usability through APIs