Applied AI Solution · Open source, research / educational
newBrain
A Turkish causal LLM built from scratch in PyTorch.
The Need & Context
Understanding modern language models deeply means building one end to end (tokenizer, architecture, training and fine-tuning) rather than only calling an API.
Solution & Execution
Implemented a causal GPT transformer and its BPE tokenizer from scratch in PyTorch, added MLX LoRA fine-tuning, and wrapped it in a FastAPI/Three.js workspace with RAG and multi-agent chat.
Key Outcomes & Deliverables
- Custom BPE tokenizer and a causal GPT transformer implemented in PyTorch (RoPE, SwiGLU, SDPA).
- LoRA fine-tuning on Apple MLX.
- FastAPI + Three.js workspace with RAG (ChromaDB) and multi-agent chat.
Capabilities & Technologies
- Python
- PyTorch
- MLX
- FastAPI
- Three.js
- ChromaDB