Hello, I'm

Samrakshyan Adhikari

I'm a |

AI Research Fellow at Algoverse • Researching ML for wildfire prediction & medical imaging at Texas State University • Building intelligent systems that make a difference.

Samrakshyan Adhikari
AI Fellow
Wildfire ML
4.0 GPA
Scroll Down

Who I Am

I'm a Computer Science student and AI researcher with a passion for solving complex real-world problems through machine learning and elegant code.

Selected globally as an Algoverse AI Research Fellow in a program directed by OpenAI and Meta researchers. Currently conducting cutting-edge research in wildfire prediction and medical imaging AI at Texas State University.

I love algorithms, books, arts, and music. My journey in tech is driven by curiosity and a desire to create meaningful impact through technology.

0 GPA
0 Research Projects
0 Full-Stack Apps

Texas State University

B.S. Computer Science (Computer Engineering) | Honors College

Expected Graduation: May 2027

President's Honor Scholarship Dean's List (All Semesters) 4.0 GPA

Research & Work

Research Intern – AI for Wildfire Prediction

Dr. Cho's Lab, Texas State University Jun 2025 – Present San Marcos, TX

Developing a deep learning pipeline for wildfire ignition risk modeling in California using multi-source remote sensing data (MODIS, GRIDMET, NLCD) and event-based cross-validation.

  • Built U-Net segmentation model for pixel-level burn probability prediction across 50+ California wildfire events (2000-2025)
  • Processed environmental data via Google Earth Engine: climate (GRIDMET), vegetation (NDVI), drought (PDSI), topography (SRTM)
  • Implemented leave-one-event-out cross-validation to test true generalization to unseen wildfires
  • Achieved interpretable probability maps with confusion visualization (TP/FP/FN analysis)
  • Scaled pipeline with Docker + GitHub Actions for reproducibility and production-readiness
U-Net PyTorch Google Earth Engine MODIS Remote Sensing Docker

Research Assistant – Osteoporosis Diagnosis

Dr. Farias Mylene, Texas State University Jan 2024 – May 2024 San Marcos, TX

Developed AI models for osteoporosis screening using facial panoramic radiography images.

  • Prepared and validated radiographic datasets with consistent metadata/annotations to reduce noise in training pipelines
  • Trained ResNet-18 CNN on 750 X-rays, achieving 94% accuracy; evaluated with F1-score, AUC, and confusion matrices
  • Applied PyTorch pipelines ensuring reproducible experiments aligned with clinical standards
ResNet-18 CNNs Medical Imaging PyTorch

Featured Projects

AI App

DisAID - Disaster Relief AI

AI-powered disaster relief platform with Google Gemini AI for intelligent emergency classification and real-time location services.

React.js Node.js Google Gemini AI Google Maps
Full Stack

E-Commerce Web Application

Full-stack MERN platform with secure authentication, product management, and REST APIs for reliable analytics.

React Node.js MongoDB Express
Web App

Interactive Bookstore

A beautiful, interactive bookstore featuring my favorite literary works with modern UI and cart functionality.

HTML CSS JavaScript Firebase

Skills & Technologies

Programming

Python
JavaScript/TS
SQL
C++/Java

ML/AI

PyTorch
TensorFlow
LLMs
Scikit-learn

Web Dev

React/Next.js
Node.js
MongoDB
PostgreSQL

Tools & Cloud

Docker
Git/GitHub
AWS/GCP
CI/CD

Let's Connect

I'm always open to discussing new projects, research collaborations, or opportunities to create impact through AI and technology.