I'm Fardeen NB

My path into artificial intelligence started from a foundation of engineering execution and entrepreneurship. In 2019, at just 18 years old, I founded FX Pattern Pro. Driven by an interest in market prediction and pattern mapping, I built and launched this desktop-based fintech application to help quantitative traders identify algorithmic pattern structures—a project that served as my first major experiment in large-scale data mapping and predictable system logic.

Realizing that the true ceiling of algorithmic logic lay inside advanced neural network frameworks, I pivoted entirely into foundational deep learning research. I pursued a Master of Engineering in Artificial Intelligence from the University of Cincinnati, specializing in the core mechanics of transformer architectures.

Today, my work bypasses superficial API integrations to focus on how models process high-dimensional information. I serve as an AI Scientist at Mercurion AI and lead the independent engineering initiative NeuralCrew. Under this banner, I achieved the complex task of pre-training and post-training a 7-billion parameter language model, Neutrino-Instruct, entirely from scratch to champion open-source data curation over raw computing scale.