Artur Loss

DeepTagger is an exceptionally valuable tool, and a critical part of ResumeCustomizer business logic. ResumeCustomizer (reumecustomizer.com) is helping jobseekers create a custom resume for every job they are applying for. Two key challenges faced during the implementation of the resume customisation idea were extracting data from customer resume uploaded on the website, and extracting keywords from the job description provided by the customer. Two seemingly very different tasks were both handled beautifully by DeepTagger. Providing several examples of resumes and marking key pieces of data to be extracted, such as name, contact details, education, and jobs, was enough to set up the system. Now all the info from customer resume is extracted from the document he or she upload and can be used to generate an improved custom CV. DeepTagger works well with PDF and DOCX documents, among other document types, which was very helpful. The second task was a little more challenging. ResumeCustomizer needed to extract keywords from the job description in order to use these in the customer resume. This requires AI to understand what word is a keyword. To complicate matters even more, ResumeCustomizer has several types of keywords. This requires actual understanding of the data and the context we are working with. Learn by example algorithm used in DeepTagger handled this task perfectly. Going through the same routine of highlighting the right words in several documents was enough to get a steady stream of keywords extracted from every job description the customer provides. Moreover, it is actually done well. Such an ambiguous task was handled perfectly. Finally, all this was connected to the DeepTagger through a convenient SDK. Now everything works quickly and efficiently. Thanks!