Divija Palleti

I am a creative, curious, and passionate person who always looks for opportunities to learn new and explore. I love to work on Front End and Back End technologies, but I like to wear many hats. I also indulged myself in machine learning and artificial intelligence during my free time, and since then, I have been passionate about creating products that could help the world in some way.
When I’m not on the job, I love to explore new things on the web and indulge my love for seeing new places.
 

I am currently pursuing a master's degree in Computer Science from the University of Massachusetts Amherst. I have a diverse background in both research and development. My thesis focuses on using generative models for explainability in black-box ML models. As a part of the European Space Agency's SOCIS program, I applied neural networks to test satellite equipment and ensure compliance with requirements. 

In the industry, I have built scalable microservices at Tekion and Interview Bit, reducing manual overhead by 40% and achieving over 99% customer satisfaction. I have over 1.5 years of experience working with various databases, including MongoDB, SQL, and DynamoDB.

In addition to my industry experience, I have completed various research projects throughout my academic career. I developed a "Neural Cryptography" project using adversarial and generative models as a new and robust security solution. I also worked at the Centre for Visual Information Technology at IIIT-Hyderabad, where I adapted Fully Convolutional DNN architectures for fully automatic, instance-level segmentation of manuscript images.

Currently a second-semester master's student at the University of Massachusetts Amherst.