Finetuned OpenAIs GPT2 : For text generation to be used by copywriting team. Built a Web Crawler : Using tools such as Selenium, BeautifulSoup and Regex. Other daily activities includes python scripting, dom scripting, data wrangling. Currently building a Web Service that would run as a cronjob to collect RSS feeds for the purpose of finetuning GPT2 on a specific topic.

Data Science for Social Good : Researched and carried out literature review for various projects in the field of traffic and transportation management and electricity conservation for the state government of Karnataka Image Classification : Worked on image datasets including but not limited to geographical and medical. Topic Modelling : A python based topic modeller that extracts top candidates for topics from an article using the Latent Drichlet Allocation. Implemented ConvNet for fake news detection

Automating Processes Using Machine Learning and Natural Language Processing Techniques Company Name Tagger : A python based project that uses multi text classification for record linkage. Company Reviews- Gibberish Classifier : A binary text classifier that classifies text as either review or gibberish. Text Summarizer : A python based extractive text summarizer that scores sentences and words based on term frequency and produces top candidates as a summary. Optimizing Email Automation using Backpropogation : A numpy based single layer perceptron network that uses Backpropogation algorithm to calculate a set of weights for given inputs of open rates of emails.