• Developed a Serverless Federated Learning library for experimenting with Federated Learning on edge devices. The paper was nominated for the Best Paper and Best Student Paper awards at IEEE eScience 2022 Conference: N. Kotsehub et al., "FLoX: Federated Learning with FaaS at the Edge," 2022 IEEE 18th International Conference on e-Science (e-Science), Salt Lake City, UT, USA, 2022, pp. 11-20, doi: 10.1109/eScience55777.2022.00016.

• Trained and deployed a Convolutional Neural Network across a network of edge devices to enable multi-staged real-time image processing. The codebase enables building serverless Edge-to-Cloud infrastructures in Python.

• Built an internal sentiment analysis and reporting tool that enables teams to assess the product's real-time performance based on social sentiment. Built with Twitter & Slack APIs, Snowflake, SQL, Hugging Face Transformers, and Python.

• Built an outlier-detection model that alerts team members via Slack notifications when a sudden drop in positive sentiment is detected.

• Automated the deployment, reporting, and alerting by using Docker and Apache Airflow.

• Grading polls and assignments for students in Multimodal Communication Course - a course about effective written and verbal communication

• Holding office hours to help out students with their assignments and confusions

• Taught 7 foundational writing skills to 40 undergraduate students at weekly sessions for a year

• Designed and facilitated writing workshops on concise and precise writing within a team of 5 other tutors

• Revised and improved lesson plans for the next year by analyzing qualitative and quantitative data from student surveys