Zahurul Islam

Led the development of high-accuracy NLP systems processing multilingual speech and text data. Deployed and optimized transformer-based models for production, significantly enhancing candidate motivation extraction.

Designed and implemented a multilingual data pipeline for automatic conversation transcription, improving accuracy by approximately 10%. Led a team of data scientists and engineers in developing scalable machine learning models.

Instructed undergraduate and graduate courses in Python and Data Science. Conducted research in NLP and machine learning, contributing to publications.

Delivered an AI model for text analysis and classification from motivation letters, enhancing recruitment efficiency. Collaborated with product managers to identify high-impact machine learning use cases.

Developed a robust multilingual tool for extracting data from textual content. Designed and maintained comprehensive machine learning pipelines.

Built a high-performing REST API for processing multilingual texts, achieving accuracy over 80%. Optimized transformer inference for production deployment.

Led a team developing machine learning models for text analysis, improving categorization accuracy by 15%. Collaborated with stakeholders to prioritize high-impact NLP use cases.

Developed multilingual tools for source and translation detection. Designed a multilingual classification tool with domain-specific accuracy exceeding 80%.