Rupesh Sapkota

Developed and integrated Fully Qualified Name Resolution API using call-graph analysis in the Python-Scalpel framework Worked on the implementation of context sensitivity for code analysis in PyCG, a Python static analysis tool for call graph generation Enhanced code quality and documentation of multiple projects; contributed to open-source libraries (Python-Scalpel, TypeEvalPy) Performed prompt engineering and fine-tuning of LLMs to assess static analysis capabilities of state-of-the-art language models

Advanced algorithm development and optimization, enhancing data analysis and AI model performance along with comprehensive code documentation. Worked on and developed end-to-end processes for complex data analysis using large datasets while significantly improving existing code infrastructures. Practical experience also includes developing recommender systems, time-series forecasting, and building predictive analysis models.

Developed a responsive website from scratch using Python and Django. Learned and used various web-development tools such as HTML, CSS, JavaScript, SQL and GoogleMapsAPI.

Developed end-to-end Knowledge Graph-based question answering system using Deep Learning and Natural Language Processing techniques with results comparable to state-of-the-art systems. Presented academic seminars on Reinforcement Learning and Secure Systems Engineering. Master’s thesis on “Evaluation of Logical and Subgraph Explanations of Node Classifiers on Knowledge Graphs” with Graph-based Machine Learning.