LT

Leo
Thakur

leo.py Real Python. Try a command.

Now

Freshman at Piedmont High School in Piedmont, California, on the software subteam of FRC Team 8033.

Four seasons of robot code

Every season, Leo has been on the software side, and every season, the robot got bigger.

  1. 1

    2023–24

    FIRST LEGO League

    Software

    MASTERPIECE season. Leo wrote the code for the team's LEGO robot.

  2. 2

    2024–25

    FIRST LEGO League

    Software

    SUBMERGED season. His second year programming the team's robot.

  3. 3

    2025–26

    FIRST Tech Challenge, Team 11201 Piedmont Pioneers

    Software lead

    DECODE season. Leo led the software team and worked on the robot's shooter and intake. The team won the Design Award.

  4. 4

    2026–27

    FIRST Robotics Competition, Team 8033

    Software subteam

    His first season on Piedmont High's FRC team, programming a full-size competition robot.

FTC 11201

Shooter

Leo worked on the robot's shooter for the DECODE game.

FTC 11201

Intake

Leo worked on the robot's intake for the DECODE game.

Design Award

FTC Team 11201 Piedmont Pioneers, 2025–26

In progress

Teaching a car to drive itself

Leo is building a self-driving RC car on the open-source DonkeyCar platform. There's no hand-written steering logic. A neural network watches the road through a camera and learns to steer by copying how Leo drives.

AI DRIVING 0.0 s

Nobody is driving

This is Leo's trained model, mypilot_cat2, completing a full lap of the Mini Monaco track in the Donkey simulator on its own. It sees only these 160×120 pixels.

The needle is the steering the network chooses, 20 times a second. It jumps between set positions because this model picks from 15 steering bins instead of outputting a smooth number.

Model steering
+0.00
Direction
Straight
23.06 sfor one full lap, driven by the model alone
35,059frames in his main simulator dataset
91training epochs for this model
40%throttle cap planned for the first real-world tests

An AI camera system

A full-stack system that watches live camera feeds, detects people, vehicles, and animals, recognizes familiar faces, and raises alerts. Leo built it with help from an AI coding assistant (Claude Code).

From camera to phone

Each frame goes through a motion check, YOLOv8 object detection, tracking, and face recognition. Events land in a database, and a FastAPI server streams everything to a desktop window, a web dashboard, and a Flutter phone app.

  • Python
  • OpenCV
  • YOLOv8
  • FastAPI
  • Flutter
~3,900lines of Python
~1,800lines of Dart
19REST API endpoints
9object classes it watches for

Toolbox

Languages

  • Pythonfavorite
  • Java
  • C++

AI and vision

Keras, TensorFlow Lite, OpenCV, YOLOv8, face_recognition, Pillow

Apps and back ends

FastAPI, SQLAlchemy, SQLite, Flutter

Robots

FIRST LEGO League, FIRST Tech Challenge, FIRST Robotics Competition, DonkeyCar

Get in touch

For robotics, internships, and summer programs.

↑↓ to move Enter to select Esc to close