17 y/o · Independent AI Researcher · India
Building artificial intelligence that goes beyond token prediction
I'm a self-taught AI researcher from Moradabad, India — no CS degree, no lab access, just curiosity and obsession.
I've spent the last year reverse-engineering how the brain works to build AI systems that are fundamentally different from today's transformers. My work spans novel architectures, training frameworks, and cognitive systems — all designed to be efficient, interpretable, and closer to how intelligence actually operates.
Replaces standard attention with Q-Compass — sequence mixing grounded in reinforcement learning navigation theory instead of geometric similarity. Three projections instead of four. 69% fewer attention parameters. One mechanism handles text, vision, audio, and world modeling.
A modular multi-agent cognitive architecture featuring 12 specialized domain experts collaborating through Web-of-Thought (WoT) reasoning.
A plug-and-play fine-tuning framework that skips samples the model already knows — routing compute to hard samples and freezing mastered ones. Up to 80% compute savings at scale.
Berry-Q0 — ~50M parameters, trained from scratch on a single laptop GPU (RTX 4050, 6GB VRAM). Text + vision, currently in GRPO reasoning training (R1-style, math domain).
The goal: push a 50M model as far as possible on reasoning. No cloud. No team. Just architecture.