Learning Journeys Early Years Planned Junior Open now Maker Open now Explorer Planned Creator Planned Researcher Planned
Get Started
  1. Early Years, journey 1
  2. Junior, journey 2
  3. Maker
  4. Explorer, journey 4
  5. Creator, journey 5
  6. Researcher, journey 6
Journey 3 of 6 · Maker

Sense. Think. Act. AI that moves something real.

Learners build an AI brain from blocks, program what it should do, and connect it to an Arduino or ESP32 board. With no kit, a robot on screen does the same.

Open nowFree during early accessChrome or Edge, nothing to installA board is optional
The Maker canvas for Gesture Lights: Record Poses, Pose Brain, Teach My AI and Try It Live in a line, with a branch to Set Pins, and the Wiring tab below showing an Arduino Uno with its built-in LED on pin 13.
Real screen, Maker: Gesture Lights, a hand gesture wired to an LED on an Arduino Uno.
  1. 1Sense
  2. 2Think
  3. 3Act
Who it is for
Learners ready to make AI do something: after Junior, or straight in with a starter project.
What learners build
Gesture lights, a hand-steered car, a sorting arm, a talking robot. 22 starters, 19 AI brains.
What you need first
Chrome or Edge with a webcam. A board is optional: an Arduino Uno or ESP32 and a few parts.
Available now?
Yes. Free during early access, with a free account.
Who it is for

The journey where an answer becomes an action

Junior ends with a model that answers. Maker adds the other half: deciding what to do with the answer, and doing it.

What makes this stage different

Maker brings three subjects together: AI, programming and electronics. Learners program with blocks in a line, if and else brackets, loops and variables, using the words real programmers use. Then a decision becomes a signal on a pin.

It also has many more kinds of AI than Junior: 19 brains for pictures, poses, sounds, faces, words, tables, sensor signals and learning by trying. Some come ready-made, most are trained by the learner, and every one can be opened with the X-Ray.

  1. Explains sense, think, act

    Describes how a robot takes in the world, decides, and acts, and which block does each part.

  2. Programs decisions

    Uses if and else, repeat, wait and variables to turn an answer into behaviour.

  3. Wires parts with care

    Follows the app’s wiring drawing for a board and its parts, and knows why motors need a driver.

  4. Compares rules and models

    Tells a rule a person wrote from a rule a model learned, and chooses which one fits.

See it

Inside the app

Real screens from Maker. Every brain has a live window, an X-Ray and a wiring view.

Try It Live with the Face Reader: the camera shows a face, the face movements are listed as bars, and the answer says Smiling, 80 percent sure. The loop at the bottom reads See, Think, Act.
Real screen. The Face Reader answering live, with The Loop at the bottom.

Watch it decide, live

The live window shows what the brain sees, its answer and how sure it is, and what the robot does about it. The Loop along the bottom lights up See, Think and Act as they happen.

How it works

Sense, think, act

Every Maker project is built from the same three moves. Blocks in a line run in order, from left to right.

  1. Sense

    A camera, a microphone or one of 19 sensor modules takes in the world: a gesture, a word, a distance, a temperature.

  2. Think

    An AI brain or a rule turns that into a decision: “open hand”, “someone is close”, “happy”. If Class and If Sure start a line only when it should.

  3. Act

    The decision drives a part: a servo turns, a motor drives, an LED lights, a buzzer sounds, a screen or a voice speaks.

Between think and act sits the program: If as a bracket with a then and an else arm, Repeat around the blocks it repeats, Wait, Stop and Remember for variables. “Download my robot’s code” writes the whole canvas out as a readable Arduino sketch.

Signature project

Gesture Car: drive a robot with your hand

The Starter projects window in Maker: Gesture Lights, Line Follower and Gesture Car at the top, then AI projects such as Gesture Gripper, Object Sorter and Talking Finder, each with its time and what it needs.
Real screen. Some of the 22 starters, with time and parts for each.

One of 22 starter projects. Learners teach two hand signals, go and stop, and the car obeys. Then they add their own: left, right, faster.

What the learner builds
A Pose Brain for two hand signals, go and stop, and a Drive block wired to two DC motors through an L298N driver.
What the AI contributes
A hand tracker finds 21 points; the Pose Brain learns which pattern is which signal.
What the learner must understand
The car does only what the brain decides, so a confused brain means a confused car. An If Sure block makes it wait until the brain is sure.
How it is tested
On the on-screen robot first, then on a real car with two DC motors and a driver board.
What learners understand

What learners can explain after Maker

  1. A decision becomes a signal

    The brain’s answer turns into a number on a pin, and the part reacts to that number.

  2. Rules and models are different tools

    A line follower uses a rule a person wrote; a gesture car uses one a model learned.

  3. Ready-made and trained brains

    Some brains arrive already trained, such as the object finder; others learn only from the learner’s examples.

  1. Confidence needs a threshold

    A robot should act only when the brain is sure enough. If Sure makes that a setting the learner chooses.

  2. Parts have limits

    Motors draw more power than a board can give, which is why a driver board sits between them.

  3. Programs run in order

    Blocks in a line run left to right, loops repeat, and a wait is time the robot spends doing nothing else.

Test and improve

Test on screen, then on the robot

Every brain is scored on examples it never saw, and the app says when it is memorising instead of learning.

Maker's Results tab for Gesture Lights: 82 percent on examples it never saw, 98 percent on the examples it learned from, a note saying big gap, it is memorising, not learning, and a table of what it mixed up.
Real screen, Maker. 82% on new examples, 98% on studied ones: the app calls the gap out.
  1. Two scores, side by side

    New examples against studied ones. A big gap means the brain memorised and needs more varied examples.

  2. What it mixed up

    A table of which class it confused with which, counted on new examples.

  3. The virtual robot first

    Programs run on the on-screen robot, so a mistake costs nothing. Then the same blocks drive the real one.

  4. Checks before running

    The app warns about two parts on one pin, a motor with no driver, or a block that ran but sent nothing.

Being tested Check a Part, a way to test one part with no blocks, is built but not yet proven on every board.

In the real world

Classroom versions of real machines

Sorting by camera

Factories sort objects by what a camera sees. Object Sorter does it with a servo arm and a ready-made object finder.

Staying on the line

Warehouse robots and lane-keeping cars follow lines. Line Follower is the classic first robot, and it works with no kit on screen.

Lights that notice you

Motion sensors switch lights and alarms. Guard Robot sounds an alarm when something comes close.

Curriculum and progression

Where Maker sits on the path

Maker follows the same project cycle as Junior and adds programming and electronics. Its 66 chapters each note where their topic sits in a school syllabus.

  1. Before

    Junior Open now

    Teach a computer to see, hear and recognise a pose.

  2. This journey

    Maker Open now

    undefined

  3. Next

    Explorer Planned

    Inside the network: layers, settings and evaluation.

For educators and parents

For teachers and parents

Start with nothing. Add parts only when the robot needs to leave the screen.

Starter projects
22 ready projects, each with its time and parts. The first ones take 5 to 10 minutes.
A textbook inside
66 chapters on the science behind every block, with questions and tap-to-see answers, in English and Hindi.
The code is visible
“Download my robot’s code” writes the canvas out as an Arduino sketch learners can read.
No kit needed to begin
The on-screen robot runs the same program, wheels and all, and a virtual world adds a line track and pretend sensors.
A first kit
Most projects need an Arduino Uno, one servo and a couple of LEDs. An ESP32 adds Bluetooth and Wi-Fi.
Phones and offline
Maker works on a phone, and keeps working offline after the first visit.
Availability and access

Open now, in Chrome or Edge

Create a free account and open Maker. Connect a board only when you want to.

Open now

Free during early access

Everything in Maker: all 19 brains, all 22 starters, every block and every chapter.

You will be asked to create a free account first if you are not logged in.

Browser
Chrome or Edge on a laptop or desktop. Connecting a board by cable uses Web Serial, which these browsers have.
Camera and microphone
For the brains that see or hear. A built-in webcam is enough.
Boards and how they connect
Proven on real boards
Arduino Uno, ESP32 and ESP32-CAM: connecting, putting the firmware on with one click, and driving parts.
Built, being tested
Arduino Nano and Mega 2560.
Cable, Bluetooth or Wi-Fi
USB on every board. On ESP32 boards, Bluetooth with no cable, or Wi-Fi joined by a six-letter pairing code.
No laptop
Bake into robot puts sensor rules onto an Uno or ESP32 so they run with the laptop unplugged. Whole lines of blocks are being tested.
Parts it can drive and read

Outputs: LEDs, RGB LEDs and LED strips, servos and 360° servos, DC motors through L298N or L293D drivers, a two- or four-wheel car, a 16-servo driver board, stepper motors, buzzers, relays, LCD and OLED screens, digit displays, an MP3 player and the laptop’s voice.

Inputs: buttons, knobs, light, distance, temperature and humidity, motion, line, touch, sound, moisture, motion and tilt sensors, a colour sensor, a spoken keyword, and 19 kinds of three- and four-pin sensor modules.

Being tested Some newer parts and screens are built but not yet proven on a board.

Questions

Questions about Maker

Do I need a robot or a board?
No. The on-screen robot runs the same program, and every starter works with no kit. Add an Arduino Uno or ESP32 when you want something real to move.
Which boards work?
Arduino Uno, ESP32 and ESP32-CAM are proven on real boards. Arduino Nano and Mega 2560 are built and being tested. The app puts its own firmware on the board with one click.
Is it the same as Junior?
It uses the same way of working, but adds programming, electronics and many more kinds of AI: 19 brains in all, from pictures and poses to words, tables, sensor signals and learning by trying.
Can learners see real code?
Yes. “Download my robot’s code” writes the whole canvas out as a readable Arduino sketch.
Does it work offline or on a phone?
Maker keeps working offline after the first visit, and has a phone layout for building and for Try It Live.
Is EagLearn Maker free?
Free during early access. There is no payment and no card.

Build one robot. Then look deeper.

Explorer, the next journey, is planned. It goes inside the networks Maker uses, and parts of it already run in Maker.