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.
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.
GEKO v0.3.1 — No-Code Web App for LLM fine-tuning.
- Browser UI — launch the full training pipeline with
geko-app, no coding needed - Live stats — loss, steps/sec, ETA, eval loss updating in real time
- One-click presets — Aggressive / Balanced / Conservative configs
- Resume from checkpoint, custom output dirs, stop/clear buttons
- Rich terminal UI — beautiful training output in the console too
→ pip install gekolib[app] · Try the live demo
| Architecture | Model | |
|---|---|---|
| Status | In Development | In Training |
| Preview |  |
 |
A new architecture + LLM combo with the potential to match GPT-3.5-level reasoning at a fraction of the compute — designed to run on consumer hardware.
Details dropping soon.