Jan 2023 – May 2023 · 4 mos · On-site

• I am working as a Course Assistant for the course COMS4111 Introduction to Databases under Professor Kenneth Ross for the spring semester.

Jan 2023 – May 2023 · 4 mos · On-site

• I am a TA for the course Deep Learning for NLP (under Prof Andrei Simion) in the Data Science Institute at Columbia University.

Aug 2021 – May 2022 · 9 mos · Delhi, India

• I worked under the supervision of Dr. Aruna Bhat on designing models to detect hate speech using twitter data. I was the team leader of our group that had three members. We worked on the entire process of collecting data, preprocessing them, and finally building models for our main task.
• Key Achievements:
• Built a dataset of 40,000 tweets with #BlackLivesMatter, #BLM, etc.
• Developed deep learning models to identify the sentimental cues to detect vitriolic comments on social media
• Co-ordinated with a group of people from linguistics and social sciences for multi-aspect annotation of tweets
• Presented the work at International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE)

Jun 2021 – Nov 2021 · 5 mos · Chapel Hill, North Carolina, United States

• As a research intern at the University of North Carolina at Chapel Hill, I worked with Dr. Snigdha Chaturvedi on de-biasing Natural Language Processing (NLP) models.We built an adversarial scrubber to protect the sensitive information of authors (such as race, age, gender, etc.) being leaked on performing a down-streaming task.We performed a statistical analysis of forty thousand tweets that were harder to de-bias.
• Frameworks used:
• PyTorch
• Hugging face transformers

Jan 2021 – May 2021 · 4 mos · Patna, Bihar, India

• I worked under the supervision of Dr. Sriparna Saha on generating extractive summaries of online streams of data. I led a team of six freshmen students in building gold standard data to compare with the model-generated summaries.

Jun 2020 – Aug 2020 · 2 mos · Sydney, New South Wales, Australia

• I was involved in research in the domain of Natural Language Processing and BCI with researchers from FEIT, University of Technology Sydney and University of Sydney (UoS), Australia.I led the team of five research student members on delivering the results efficiently to our supervisor.I co-ordinated with them on matters of task allocation and conducted weekly meetings to check on the progress of the work.We developed baseline models for early detection of Alzheimer's disease using Nepalese speech transcripts.
• Key Achievements:
• Presented the implementation of work at the International Conference on Neural Information Processing (ICONIP).
• Extended the baseline version of the models by building four deep learning models
• Achieved F1-scores 5% more than the baseline ones
• Published the extended version of baseline implementation at the International Journal of Human-Computer Studies.