- 🎓 B.Tech CSE Student @ VIT Chennai
- 🧠 Focused on Deep Learning, Computer Vision & AI Research
- ⚙️ Building scalable Full-Stack + ML systems
- 📈 Preparing seriously for DSA & Technical Interviews
- 🔬 I build systems with metrics, experiments & explainability
ISIC 2019 • CNN + Transformer Hybrid • GAN Rebalancing • XAI
🔹 Preprocessing: Hair Removal, CLAHE, Color Normalization
🔹 Binary Model: DenseNet & EfficientNet
🔹 Multi-Class Hybrid: CNN + Vision Transformer
🔹 GAN-based malignant augmentation (Training only)
🔹 Explainability: Grad-CAM & Attention Maps
🔹 Metrics: Accuracy, AUROC, Precision, Recall, F1
🧪 Designed with publication-level experimental comparisons.
Tech Stack: Next.js • tRPC • Prisma • Tailwind • LangChain • Gemini • RAG
✨ Features:
- Commit summarization
- Natural language repository Q&A
- Retrieval-Augmented Generation
- Auth + Token Gating
- Scalable production-style backend
Built as a real SaaS architecture, not just a demo app.
Tech: ESP8266 • Force Sensor • Blynk IoT
- Real-time impact force monitoring
- Mobile app integration
- Low-cost alternative to ultra-edge detection
- Demonstrated at IAIC 2025
Java Python C/C++ JavaScript SQL
React Next.js Node.js Tailwind CSS
PyTorch TensorFlow OpenCV Scikit-Learn
GANs Vision Transformers Grad-CAM
AWS (EC2, S3, Lambda)
Git GitHub VS Code Google Colab
📧 bhavesh71.logu@gmail.com
🔗 GitHub: https://github.com/Bhavesh71
Build. Break. Improve. Repeat.
Depth over hype.
Systems over templates.