
Kacper Zmorski developed an end-to-end LoRA fine-tuning demonstration notebook for online banking queries, contributing to the IBM/watsonx-ai-samples repository. He designed the solution to be cloud-ready and reproducible, guiding users through environment setup, data preparation, fine-tuning, and model evaluation. Using Python, Jupyter Notebooks, and machine learning techniques, Kacper addressed practical challenges in customizing foundation models for enterprise onboarding scenarios. The notebook’s comprehensive documentation and production-oriented workflow enable customers to experiment with LoRA-based model adaptation in a realistic banking context. His work deepened the sample portfolio and improved reproducibility for AI model deployment in regulated industries.

Performance-focused month (2025-08) delivering an end-to-end LoRA fine-tuning demonstration for online banking queries. The work enhances enterprise onboarding by providing a cloud-ready, reproducible sample that showcases practical model customization, evaluation, and deployment considerations. No critical bugs were reported in this scope.
Performance-focused month (2025-08) delivering an end-to-end LoRA fine-tuning demonstration for online banking queries. The work enhances enterprise onboarding by providing a cloud-ready, reproducible sample that showcases practical model customization, evaluation, and deployment considerations. No critical bugs were reported in this scope.
Overview of all repositories you've contributed to across your timeline