Blog

Notes on NLP, ML systems, and research

Introducing synthetictext

synthetictext is an LLM-powered Python package for generating synthetic text data for text classification tasks.

2 min read · June 01, 2026
How we won the PAN 2024 Text Classification Challenge

How our IUCL team won PAN 2024 opposition stance detection with data augmentation and fine-tuned transformers.

2 min read · July 31, 2024
Getting started with Selenium Webdriver and Requests in Python

Scrape the USPTO patent database with Selenium WebDriver and download PDFs with Requests.

12 min read · June 13, 2024
Problems and Nuisances

Problems vs nuisances in product management - which user issues are worth solving.

2 min read · January 15, 2021