|
Week - 1 |
Informing about Internship II procedures, advanced technical expectations, and R&D-oriented internship regulations. |
|
Week - 2 |
Examination of the host institution's large-scale data architecture, AI production pipelines, and computing infrastructure (GPU/Cloud systems). |
|
Week - 3 |
Adaptation to the institution's advanced AI project management tools, version control standards (Git workflows), and CI/CD processes. |
|
Week - 4 |
Definition of the complex industrial problem to be solved, literature review, and advanced architectural requirement analysis within the scope of the internship. |
|
Week - 5 |
Secure extraction of large-scale or complex (visual, textual, sensor, etc.) datasets from corporate databases or data lakes and establishing data pipelines. |
|
Week - 6 |
Advanced data preprocessing, data augmentation, noise reduction, and structuring processes for complex datasets. |
|
Week - 7 |
Architectural design of modern deep learning (CNN, RNN, LSTM), natural language processing, or reinforcement learning models based on the problem type. |
|
Week - 8 |
Training developed advanced models using transfer learning or custom architectures on corporate GPU/Cloud infrastructure. |
|
Week - 9 |
Implementing methods to prevent overfitting (Regularization, Dropout), hyperparameter tuning, and gradient optimization processes in deep models. |
|
Week - 10 |
Analysis and error investigation of trained advanced models based on complexity, accuracy, memory footprint, and inference time criteria. |
|
Week - 11 |
Planning and simulation of deployment/MLOps processes, API integrations, or microservice architectures for the developed models. |
|
Week - 12 |
Integration of the solution with industrial systems, scalability analyses, and completion of corporate technical documentation. |
|
Week - 13 |
Reporting of all advanced engineering and modeling activities conducted during the internship in accordance with academic rules. |
|
Week - 14 |
Submission of the Internship II Report containing advanced technical achievements to the committee and completion of course evaluation processes. |