Pi-treeAI that reads with you

Why Pi-tree? ​

Most AI tools help you skip past material. That works when you already understand the domain. When you don't, skipping is exactly the problem. Pi-tree treats reading as a process worth having — one that expands what you're capable of understanding.

What a reading session looks like ​

📖 Reading: Thinking, Fast and Slow (Kahneman)

Root
├── What is System 1 vs System 2?
│   ├── How does this relate to cognitive biases?
│   │   └── Anchoring bias deep-dive
│   └── Real-world examples in decision making
├── Chapter 3: The Lazy Controller
│   └── Why do we avoid effortful thinking?
└── Comparison with Nassim Taleb's ideas
    ├── Black Swan connection
    └── Antifragility and heuristics

Each node is a conversation branch with full context. Go deep on any concept, then navigate back to explore something else — no context lost.

Why trees work better for LLMs ​

The tree structure isn't just a UX choice — it makes the AI more accurate and cheaper to run.

In a linear chat, every message is packed into the context window. After 30 turns spanning three topics, the model hallucinates, loses the thread, or ignores your latest question. Trees fix this at the architecture level:

• Focused context — Each branch carries only its path from root to current node. Less noise → more accurate responses.
• Token savings — A 50-message linear chat sends all 50 every turn. A tree with 5 branches of 10 sends only ~10. Lower cost, faster responses.
• Less hallucination — Context pollution is a primary cause of hallucination in long conversations. Isolated branches keep the model grounded.
• Longer effective conversations — Linear chats degrade well before the context limit. Trees keep each branch short, so quality stays high across hundreds of messages.