High school student independently learning machine learning. I build end-to-end ML systems—from raw data to deployed models—with a focus on real-world impact.
// Sakib · BD
I'm a high school student from Bangladesh, self-teaching machine learning with an end-to-end mindset — not just training models, but deploying them. I've built systems for astronomical classification, seismic data analysis, and exoplanet detection, and competed in NASA's Space Apps hackathon.
I'm still early in this journey — learning continuously, not from courses, but through books, documentation, and building things that actually run. The gap between theory and deployment is where I spend most of my time. On my GitHub, you'll also find repositories where I openly share code and experiments from whatever I'm currently reading or working through.
Alongside ML, I run Curiosity-Code — a personal collection of mathematical experiments and utility scripts.
Astronomical object classification system trained on sky survey data. Differentiates stars, galaxies, and quasars with high precision — end-to-end, from preprocessing to deployed API.
↗ 02Seismic data analysis pipeline for detecting and characterizing earthquake signals. Built to process real waveform data, extract features, and classify seismic events with strong accuracy.
↗ 03Exoplanet detection model built for NASA's Space Apps 2025 hackathon. Uses a stacking ensemble to identify planetary transits in stellar light curves — combining multiple learners for robust signal detection.
↗ 04Urban data analysis project exploring patterns in city-level datasets. Applies machine learning to surface meaningful structure in complex, multi-dimensional urban environments.
↗ 05A personal collection of mathematical experiments and utility scripts — built for curiosity, not a portfolio. Covers π approximation (Leibniz vs Ramanujan convergence), Euler's number via limit definition, chord geometry, coordinate tools, the golden ratio φ, and more. All Python, all for the joy of it.
↗Whether it's about ML, a project, or just to say hello — I'm always happy to hear from you.