# Aasman Bashyal

> Research & AI Architect at Angelswing, Remote.

## Complete text

- [Full portfolio Markdown](https://aasmanbashyal.com.np/portfolio-full.md): Full biography, resume, every project and article, and gallery image links in one Markdown file.

## Public portfolio

- [Portfolio Markdown](https://aasmanbashyal.com.np/portfolio.md): Public profile, work index, and contact links.
- [About Markdown](https://aasmanbashyal.com.np/about.md): Complete biography.
- [Resume Markdown](https://aasmanbashyal.com.np/resume.md): Full work history, education, awards, and skills.
- [Gallery Markdown](https://aasmanbashyal.com.np/gallery.md): Image titles and original image links.
- [Text versions](https://aasmanbashyal.com.np/agents/): Index of the public text files.
- [Resume PDF](https://aasmanbashyal.com.np/pdf/resume_aasman_20261023.pdf): Published resume PDF.

## Articles

- [Automatic Modulation Classifier](https://aasmanbashyal.com.np/projects/amc.md): Identifying signal modulation types through automatic classification.
- [Robotics Projects](https://aasmanbashyal.com.np/projects/robotics.md): Robots our team built for ABU Robocon.
- [Simultaneous Localization and Mapping (SLAM)](https://aasmanbashyal.com.np/projects/slam.md): Robot localization and mapping using a grid map and particle filter.
- [Gaussian Splatting](https://aasmanbashyal.com.np/projects/gaussian.md): Rendering novel views with Gaussian Splatting, with the mathematics behind the technique.
- [Structure from Motion (SfM)](https://aasmanbashyal.com.np/projects/sfm.md): An SfM pipeline covering feature detection, camera pose estimation, triangulation, and 3D reconstruction.
- [Automatic Polygon Drawing](https://aasmanbashyal.com.np/projects/polygon.md): Drawing polygon masks automatically with the Segment Anything Model.
- [3D Terrain Editing](https://aasmanbashyal.com.np/projects/terrain.md): Editing 3D terrain and flattening selected surfaces into 2D while preserving texture.
- [CAD Snapping](https://aasmanbashyal.com.np/projects/snapping.md): Snapping for precise measurements in a CAD overlay.
- [When weight pruning helps TinyML](https://aasmanbashyal.com.np/blogs/tinyml.md): How weight pruning shapes TinyML models, and where hardware constraints limit the gains.
- [When low-pass filtering helps graph neural networks](https://aasmanbashyal.com.np/blogs/spectral.md): Why low-pass filtering helps graph neural networks, and when removing high-frequency signals can lose useful information.