- 💬 Ask me about .NET, Vue, React, Flutter, Sql, LLMs — and lately ESP32-S3, C++, Arduino, robot hardware
- 🔭 I'm currently working on a Wi-Fi remote-control and auto-avoidance car built on ESP32-S3
- 🌱 I'm currently learning robotics from zero electronics: a 6-month path to ROS2, control theory and LeRobot
I started with zero electronics background and am working through a 6-month path to job-ready robotics: electronics → microcontrollers → mechanics → ROS2 → control theory → LeRobot.
- 📍 Month 1 (of 6): electronics fundamentals, breadboarding, soldering, ESP32-S3, sensors, DC motors
- 🔧 Current build: an ESP32-S3 car you drive from a phone browser, with one button that hands it over to ultrasonic auto-avoidance
- 🎯 Next: microcontrollers in depth — motor drivers, servo/stepper control, IMU and ultrasonic ranging; then mechanical design and 3D printing; then ROS2 with Gazebo simulation; then PID tuning and Kalman filtering
| Month | Focus |
|---|---|
| 1 | Electronics fundamentals, breadboard, soldering, ESP32-S3, sensors, DC motors |
| 2 | Microcontrollers in depth, motor drivers, servo/stepper control, IMU, ultrasonic ranging |
| 3 | Mechanical design, CAD, 3D printing, laser cutting, chassis building |
| 4 | ROS2 simulation, Gazebo, Nav2, enterprise-level practice |
| 5 | Control theory, PID tuning, Kalman filtering, maths foundations |
| 6 | LeRobot, reinforcement learning for robots, specialisation, portfolio and job hunting |
Everything lives in robotics-engineer-learning-tutorial — one folder per day, each with the hardware list, wiring, code and measured data.
- 🤖 Wi-Fi remote-control + auto-avoidance car (month 1, in progress) — ESP32-S3, TB6612FNG, two TT motors, HC-SR04 sonar. Driven from a phone browser; one button hands it over to ultrasonic auto-avoidance, and the watchdog guards the mode, not just the direction
- ⚙️ Actuator & sensor bench rig (month 2, mostly done alongside the car) — motor driver, servo, IMU and ultrasonic ranging, each verified standalone on the bench before anything went on the car
- 🛠️ 3D-printed chassis (month 3, planned) — CAD design, 3D printing and laser cutting, replacing the hand-wired frame
- 🧭 ROS2 + Nav2 on a simulated car (month 4, planned) — the same car rebuilt in Gazebo, then driven through the Nav2 stack
- 🎛️ PID + Kalman filtering on the real car (month 5, planned) — tuning and filtering applied to the physical chassis, not just in simulation
- 🤖 LeRobot project + portfolio (month 6, planned) — reinforcement learning for robots, plus the portfolio for job hunting
I hold every entry to the same five things: what I did, the code, the test results, what went wrong & how I fixed it, and references.
"What went wrong & how I fixed it" is the part that matters most. My rule for it: never write "spent a long time debugging" — state the cause, how it was localised, what changed, and how the fix was verified.
Worked examples live in the day folders.