<aside> 👋 I am a final year undergrad at the Department of Computer Science & Engineering, IIT Delhi. After undergrad, I aim to pursue higher studies and build on my research experience.

</aside>

               [Email](<mailto:[email protected]>) / [CV](<https://drive.google.com/file/d/1PU8aUxpOvYkKMeaii7_-uCjHbXY5k8zB/view?usp=sharing>) / [Github](<https://github.com/amangupt01>) / [Linkedin](<https://www.linkedin.com/in/amangupta27/>)

Research Interests

I am broadly interested in Machine Learning, Computer Vision & Systems. Specifically, I have worked on Object Detection, Medical Image Analysis, Satellite Imagery, and on ML systems.

Publications

• **Ultra-high resolution, multi-scale, context-aware approach for detection of small cancers on mammography.** Krithika Rangarajan, Aman Gupta, Saptarshi Dasgupta, Uday Marri, Arun Kumar Gupta, Smriti Hari, Subhashis Banerjee & Chetan Arora, Scientific Reports - Nature Journal 2022
• Tracking Socio-economic Development in Rural India Using Satellite Imagery. Anant Lunia, Akshay Sarashetti, Aman Gupta, Aaditeshwar Seth. ACM Compass’23 (yet to be published)

Research Experience

I have been part of the following Research Groups:

• **ACT4D: Appropriate Computing Technologies for Development** January 2021 - Present

  • Supervisor: Prof. Aaditeshwar Seth
  • Field: Machine Learning & Computer Vision
  • Description: ****Implemented two-staged CNN & regression pipeline to predict village development indexes at the village level. Executed occlusion studies to analyze the interpretability of CNN weights. Performed Image Segmentation to find land use patterns of villages and their correlation with development. [Github]

• **Computer Vision Group** August 2021 - May 2022

  • Supervisor: Prof. Chetan Arora & Dr. Krithika Rangarajan
  • Field: Computer Vision
  • Description: Trained YOLOv5 pipeline to detect and localize tumor masses in mammography. Experimented with various data augmentation techniques to improve the accuracy of small tumor masses detection. [Github]

• RobustML Research Group April 2022 - Present

  • Supervisor: Prof. Abhilash Jindal
  • Field: Operating Systems, Parallel & Distributed Programming
  • Description: Group aims to develop a Conjecture and Refutation Programming prototype Popper that makes writing ML workflows efficient. I worked on applying liveness analysis to remove dead nodes from task DAGs. Executed Distributed Tracing to identify possible bottlenecks during Runtime. Improved the design of aggregators to help reduce runtime by 2x. Currently working on efficiently implementing video processing workflows.

  Summer Internships

  • **Microsoft, Hyderabad** June 2022 - July 2022
    • I worked with the Azure Cloud Compute Team at Microsoft. Worked involved improving the restore pipeline of blob-to-blob transfers in SaaS infrastructure. I designed and implemented batch downloads with added support for scalability, fault tolerance, and crash recoverability. [Presentation]
  • **Limechat, Bangalore** June 2021 - July 2022
    • Description: Analysed chatbot-user conversations to improve the intent-entity detection ability of the NLU model by data augmentations. Designed ways to enable clients to adjust bot responses for specific customer requests
  • **Nova Benefits, Bangalore** June 2020 - July 2020
    • Implemented Graphql queries & mutations, Postgres tables structures, and APIs to sync & backup client files on GCP