🧪 TestingAI — Experimental AI Sandbox & Model Benchmarks

"My private laboratory for testing prompts, benchmarking fine-tuned weights, and prototyping experimental AI architectures."
Maintained by Teja Priyan

---

🔬 About This Repository

Welcome to my AI workbench! I'm Teja Priyan.

Before deploying any model (like my 8B SQL reasoning model or computer vision pipelines) into production, I test, experiment, and break things here first.

This repository serves as my living playground for:

• 📊 Model Benchmarking: Testing reasoning accuracy, latency, and context recall across open-source LLMs.
• 🎯 Prompt Engineering & Evaluations: Experimenting with zero-shot, few-shot, and chain-of-thought system prompts.
• 🧠 Fine-Tuning Experiments: Early scripts for LoRA, QLoRA, and GRPO reinforcement learning runs.
• 🛠️ Synthetic Data Pipelines: Generating and verifying domain-specific training pairs.

---

🗂️ Lab Structure & Ongoing Experiments

TestingAI/
├── benchmarks/      # Automated accuracy and latency evaluation scripts
├── prompts/         # Curated system prompts and chain-of-thought templates
├── prototypes/      # Small Python test scripts and model inference tests
└── tests/           # Unit evaluations for SQL generation and formatting

---

⚡ Quick Start

# Clone the repository
git clone https://github.com/TejaPriyan/TestingAI.git
cd TestingAI

# Set up a virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: .\venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt (when added)

---

📄 License

This repository is open-source under the MIT License © 2026 Teja Priyan.