Interested in Participating GSoC Contribution

[Samia Haque]

Hello NumFocus Community,

I am Samia Haque, a recent graduate with a B.Sc. in Software Engineering (major in Data Science). I have been actively working on integrating computer vision with post-hoc interpretability techniques and LLM-based automation, exploring how these technologies can enhance decision-making in various domains.

My Experience & Skills

• Object Detection & Post-hoc AI Explainability: I am currently working on multi-stage traffic anomaly detection, where I utilized YOLOv9 and YOLOv10 to detect traffic-congested regions from images. I am also working on its post-hoc interpretability and LLM integration. You can view my work on this project here: Deep Learning for Explainable Traffic Anomaly Detection in Dhaka.
• LLM Integration & AI Automation: I have experience in prompt engineering, dataset curation, and incorporating human feedback to improve LLM outputs. While I am not deeply involved in model fine-tuning or RLHF at a model-training level, my work in structuring datasets and evaluating outputs supports effective AI-driven automation.
• Contributions to DeepForest: I have explored the DeepForest codebase and contributed by submitting a pull request on a good first issue, along with some documentation fixes. This experience has given me hands-on insight into the codebase and its workflows.

I am particularly interested in contributing to the Data Retriever project. This project combines object detection with AI-driven automation, and it aligns with my interest in developing interpretable and adaptable AI systems.
I am committed to dedicating the necessary time to contribute and learn from this project. Here are several ways to reach me:

GitHub: Samia35-2973
LinkedIn: Samia Haque
Email: samiat...@gmail.com
Discord Username: samia_tisha