Every team has a story waiting to be told. Our story is about what it takes to build an original proofreading solution. Actually, it is a bunch of stories of people, technologies, and day-to-day decisions. And today we feel like sharing them broadly.
So we created Beyond WebSpellChecker – a Medium blog with our findings and behind-the-scenes insights in software development at large and in natural languages processing, in particular. We hope it might prove useful for our colleagues in the industry, who move in the same direction and face similar tasks.
In our first piece we contemplate of the promising trends in proofreading technologies and on our approach at getting there. Check out the Multi-headed model based on BERT to solve Grammatical Error Correction tasks more efficiently, providing an overview of the technology we presented at this year’s conference of the Association for Computational Linguistics in Florence.
Please clap to the stories you find helpful and follow Beyond WebSpellChecker not to miss the new posts.
As for the ongoing communication on the products and the company updates – we will keep posting them here. Keep in touch!