Add demo script with clear instructions

Author: haasonsaas

run_demo.sh

@@ -0,0 +1,56 @@
+#!/bin/bash
+set -e
+
+echo "╔════════════════════════════════════════════════════════════╗"
+echo "║   Tinker Evaluation-Driven Fine-Tuning Demo               ║"
+echo "╚════════════════════════════════════════════════════════════╝"
+echo ""
+
+if [ -z "$TINKER_API_KEY" ]; then
+    echo "❌ Error: TINKER_API_KEY environment variable not set"
+    echo ""
+    echo "Please set your Tinker API key:"
+    echo "  export TINKER_API_KEY=sk-your-key-here"
+    echo ""
+    exit 1
+fi
+
+echo "✓ Tinker API key found"
+echo ""
+
+if [ ! -d ".venv" ]; then
+    echo "📦 Creating virtual environment..."
+    python3 -m venv .venv
+fi
+
+echo "📦 Installing dependencies..."
+source .venv/bin/activate
+pip install -q -r requirements.txt
+
+echo "✓ Dependencies installed"
+echo ""
+
+echo "🚀 Starting evaluation-driven training loop..."
+echo "   - Base model: meta-llama/Llama-3.1-8B-Instruct"
+echo "   - Training data: 20 examples (demo_data.jsonl)"
+echo "   - Max rounds: 3"
+echo "   - Eval threshold: 0.75"
+echo "   - Initial LR: 0.0003 (decays by 0.6x per round)"
+echo ""
+
+python trainer_with_eval.py --config demo_config.json
+
+echo ""
+echo "✅ Demo complete!"
+echo ""
+echo "What happened:"
+echo "  1. Loaded 20 training examples"
+echo "  2. Fine-tuned model with LoRA on Tinker infrastructure"
+echo "  3. Evaluated model on QA tasks"
+echo "  4. If score < 0.75: adjusted LR and started Round 2"
+echo "  5. Repeated until threshold met or max rounds reached"
+echo ""
+echo "Next steps:"
+echo "  - Review checkpoints saved in Tinker"
+echo "  - Enable EvalOps integration to track all runs"
+echo "  - Replace simple_eval.py with real Inspect AI tasks"