I am a Data Scientist at Feedzai, where I work on transaction and loan application fraud models. I love web scraping, Python, NLP and being the go-to guy when data holds the key to solving a problem.
I studied computer science and finished my graduate degree in data science in January 2021. Discovering my love with data since undergrad, I interned with multiple industries working on database management, query optimization and data analysis. Since mid-2020 I have been working with ML/DL at organizations and personal side-projects.
In my spare time, you can find me doing HIIT, reading, cooking or trying to perfect latte art!
Links:
GitHub | LinkedIn | Twitter | Email
Chemex Diffused: Fine-tuned Stable Diffusion on Chemex coffee brewer images to implant in fun output domain of the model. My favorite one is “A photo of cmexsks drink with ice cream on top”. Give a try:
Corny AI: Prompt engineered GPT3 to write corny jokes from any user question. Try it out yourself:
Spanish Reviews Rating: Electra model (discriminator) fine-tuned on Spanish movie reviews (muchocine) dataset to predict star rating
shahp7575/electricidad-base-muchocine-finetuned · Hugging Face
GPT2-Horoscopes: A GPT-2 model fine-tuned on one year of horoscope data, writes horoscopes based on a chosen category. Try the demo:
Modern Slavery Statements Analysis: Part-of-speech and subject-object syntactic representation analysis in an attempt to produce additional insights on the modern slavery statements dataset. My analysis notebook:
LaLiga 2018-19: From data collection to an EDA dashboard summarizing match results and statistics for FC Barcelona's 2018-19 season. Have a look:
Web Scraping: I also scrape interesting datasets and open-source them on Kaggle.