Hi there! I'm Aashwika — a curious mind deeply invested in AI, real-world problem-solving, and systems that scale. I enjoy leading tech initiatives, exploring user-first thinking, and turning complexity into clarity.

Currently diving into:

• 💻 Applied AI systems and model deployment
• 🧠 Multimodal deep learning for humanitarian applications
• 🎯 Building towards solving real-world problems using AI
• 🌍 Regional AI deployment for disaster response (Telangana, India)

🌐 MMDRS — MultiModal Disaster Response System End-to-end humanitarian AI platform fusing social media text and imagery via BERT + ResNet-50 dual-encoder fusion architecture. Four-head multi-task classifier predicting disaster type, severity, humanitarian impact and informativeness simultaneously. Features cost-sensitive Expected Operational Risk (EOR) scoring, semantic dissonance detection, and real-time geospatial dashboard. Deployed at Telangana scale with NGO partners. 82–85% accuracy, 276ms inference latency.

🌡️ Heart Disease Risk Prediction Built an ML pipeline using Random Forest, Logistic Regression, and Gradient Boosting to predict heart disease risk — complete with accuracy comparisons & crisp data visuals.

🦺 PPE Detection using YOLOv5 Built a real-time safety system detecting helmets and vests in industrial environments. Vision meets vigilance.

🧠 IdeaBank AI-first web app for capturing and refining ideas in real-time, powered by Firebase, Gemini API, and Google IDX. Designed for the future of ideation.

📺 Intel Unnati DL Streamer Project Designed a smart video analytics pipeline using DL Streamer & OpenVINO on Intel's platform. Performance redefined.

• 🌍 MMDRS regional expansion — Andhra Pradesh, Maharashtra, Karnataka
• 🧠 Cross-modal attention heatmaps for explainable disaster AI
• 📊 CI/CD pipelines for model deployment
• 🤝 Low-code internal tools for idea management

💻🎯✨🚀📊☕💡🐛🌠🌌✨🌙🌟🚀💫🛰️🔭🪐🎨🧵📸🧠💭✨🧩💡🖌️🗯️🐾☕💻🧣🧠✨📚🧁🌟🔭