At WaveAccess, I developed an end-to-end Retrieval-Augmented Generation system for searching and answering questions from PDF documents. I built preprocessing and chunking pipelines, integrated Qdrant vector search, and used large language models to generate answers from retrieved context.
I also automated downloading PDF documents from Excel metadata using Python and REST APIs, with retry logic, logging, and tests. I refactored the project into a Python package using Poetry and applied tools including Git, Ruff, and MyPy.
At MG Group, I use SQL and 1C to analyze supermarket sales and inventory data, identify discrepancies, and evaluate product performance. I also create Tableau dashboards and use Jira to track tasks for a project supporting communication and task assignment between management and branch directors.
