class Priyanka:
name = "Priyanka Gounder"
degree = "BCA — Silver Oak University"
interests = ["AI/ML", "Edge AI", "LLMs & RAG", "Full-Stack Dev"]
tools = ["Python", "TensorFlow", "LangChain", "Streamlit", "OpenCV"]
currently = "Exploring LangChain · RAG Pipelines · LLM Integration"
fun_fact = "I built a defect detector that runs 100% offline on a chip 🔬"
🔬 Real-time sensor defect detection — completely offline, zero cloud dependency
| Feature | Detail |
|---|---|
| 📷 Hardware | ESP32-CAM — captures live sensor images |
| 🧠 Model | Edge Impulse image classification — 99.6% confidence |
| 🎯 Detects | DHT11, Ultrasonic, IR Sensor, Battery Cell — Normal / Damaged / Cracked |
| 🌐 Dashboard | Live WiFi web UI — auto-refreshes every 2 seconds |
| 🏭 Use Cases | Electronics manufacturing · Automotive · Medical devices · PCB QC |
🤚 Control LED & Fan without physical touch — contactless home automation
- Real-time hand landmark detection using OpenCV + MediaPipe
- Gesture recognition mapped to ON/OFF device commands via webcam
- Demo video published on LinkedIn 🎥
🗂️ Upload any PDF → Get summaries & ask questions instantly
- Clean Streamlit UI with AI-powered summarization & Q&A
- Focused on API integration and backend logic
💼 Fully responsive with light/dark mode, smooth animations & project showcase
