Research of potential features for an AI Explainability platform, on topics such as feature correlation, gradient estimation, and data poisoning.

Python developer of back-end and ML solutions across several external and internal projects. Highlights:- Researched and implemented AI Explainability techniques for an XAI platform, from POC to production. Topics covered include feature importance, adversarial robustness and outlier detection.- Developed end-to-end NLP solutions and ensured data quality for a large-scale webscraping pipeline, enabling the identification of business descriptions from company sitemaps.- Implemented component logic for the ingestion, analysis, and visualization of key corporate financial metrics.- Gained hands-on experience with other ML topics such as named entity recognition, multi-task learning, and pre-GPT language models and vector embedding techniques.

Implemented traditional ML algorithms for a proposed social network for academics; worked on attribute extraction and subject identification from scientific articles in PDF format, and a recommendation system for users with similar interests and papers.