Turning raw data into meaningful insight. I build machine learning models, analyze complex datasets, and communicate findings that drive decisions.
Hi, I'm Biraj Rijal, a machine learning enthusiast from Kathmandu, Nepal. I'm passionate about applying machine learning and statistical analysis to solve real-world problems.
Currently studying and building projects across audio signal processing, NLP, and predictive modeling. I believe data tells a story — my job is to find it and communicate it clearly.
When I'm not training models, I'm exploring new datasets, writing about what I learn, or contributing to open-source projects.
Extracted MFCC features from audio signals and trained a Linear SVM classifier to distinguish between audio classes with high accuracy.
Built and tuned a Random Forest model for structured tabular data, achieving strong predictive performance through careful feature engineering.
Implemented and experimented with deep learning architectures for image and text classification tasks using modern frameworks.
A deep dive into Mel-Frequency Cepstral Coefficients and how they're used to represent audio signals for machine learning tasks.
Despite the rise of deep learning, ensemble tree methods continue to dominate structured data problems. Here's why.
Resources, communities, and paths for aspiring data scientists in Nepal — what worked for me and what I'd do differently.
I'm open to collaborations, research opportunities, and interesting data problems. Feel free to reach out!
hello@birajrijal.com.np