Building intelligent systems that bridge machine learning with real-world deployment constraints.
🎓 Electronics & Communication Engineering Graduate (VTU)
📍 Bangalore, India
🧠 Focused on Applied Machine Learning, Computer Vision, Robotics, and Intelligent Embedded Systems
⚙️ Building real-time systems that combine ML, hardware, sensors, and deployment-oriented workflows
📌 Interested in ML Engineer, Computer Vision, Robotics-AI, and Embedded AI roles
📑 Click to view Table of Contents
- 🧠 Applied Machine Learning: Feature engineering, classification systems, real-time inference, and human activity recognition
- 👁️ Computer Vision: Pose estimation, movement validation, and visual feedback pipelines
- 🔌 Embedded Systems & IoT: ESP32 systems, IMU sensing, wireless communication, and real-time data streaming
- ⚡ Intelligent System Integration: Combining ML, hardware, embedded communication, and UI workflows into deployable systems
| Category | Technologies |
|---|---|
| Programming | Python, C, C++, SQL |
| Machine Learning | Random Forest, Classification Models, Feature Engineering, Model Evaluation |
| Libraries & Frameworks | NumPy, Pandas, Scikit-learn, OpenCV, MediaPipe, FastAPI, Streamlit |
| Embedded & Systems | ESP32, IoT Architectures, Sensor Fusion, Real-Time Systems, Wireless Communication |
| Tools | Git, GitHub, Jupyter Notebook, VS Code, Arduino IDE |
- Edge AI
- Computer Vision
- Robotics Automation
- Human Activity Recognition
- Intelligent Embedded Systems
- Real-Time ML Systems
Tech Stack: Python, Scikit-learn, MediaPipe, OpenCV, ESP32, Streamlit
- Developed a multi-modal rehabilitation monitoring system combining IMU sensor fusion and computer vision-based pose estimation
- Built Random Forest models achieving 98.7% correctness, 96.4% exercise classification, and 94% posture accuracy
- Engineered feature pipelines using statistical and frequency-domain analysis for human activity recognition
- Designed resilient fallback workflows supporting independent IMU-only and vision-only inference modes
- Built real-time dashboard for monitoring, inference visualization, and corrective feedback generation
Tech Stack: ESP32, PCA9685, C++, I2C, PWM, Embedded Systems
- Built a WiFi-controlled 6DOF robotic arm with browser-based calibration interface
- Implemented calibration-driven pick-and-place execution pipeline for repeatable autonomous operation
- Solved embedded hardware challenges involving servo jitter, voltage instability, and I2C communication failures
- Designed the system to operate independently after calibration without continuous manual input
Tech Stack: Python, Scikit-learn, XGBoost, SHAP, Pandas, NumPy
- Developed an end-to-end machine learning pipeline for diabetes risk prediction using clinical health parameters
- Compared Logistic Regression, Random Forest, and XGBoost models using Accuracy, F1 Score, and ROC-AUC evaluation metrics
- Achieved ~0.83 ROC-AUC using Random Forest classification with preprocessing and feature evaluation workflows
- Implemented SHAP-based explainability pipeline for feature importance visualization and model interpretability analysis
- Built modular training and explainability workflows supporting reproducible experimentation and evaluation
Tech Stack: ESP32, RTC Module, Embedded C++, IoT
- Built IoT-based automated medication dispensing and adherence tracking system
- Implemented RTC-based scheduling and real-time intake logging workflows
- Designed modular architecture supporting future healthcare monitoring integrations
- 🥈 2nd Prize - National Level Project Expo (46 teams)
- 👩💼 Lead Coordinator - IEEE Technovate 2K25 (700+ participants)
- 🎯 Coordinator - ISTE Technisum (500+ attendees)
- 📧 Email: jahnavidade@gmail.com
- 🔗 LinkedIn: linkedin.com/in/jahnavi-dade
- 💻 GitHub: github.com/DadeJahnavi