Skip to content
Home/Australia/Western Australia/Perth/Midland
Midland, City of Swan, Western Australia

AI and programming classes in Midland

What can Midland learners gain from AI and programming classes?

The ability to build machine learning models and to measure them honestly, keeping data used for choosing a model apart from data used for judging it. Young children start in Scratch at six, and older learners build and test models in Python. Midland had 6,335 residents in 2021.

Around Midland station, about one stop record in five carries a Y in the state file's bay field. Try 852 simple rules to predict which, keep the one that scores highest on your test data, and it looks like progress. Score that same rule on data it has never met and almost all of the progress vanishes.

Book the free classBook a priority demoAsk on WhatsApp

Facts last verified 13 October 2026. Teaching is online; no Midland branch is claimed.

Course picks for Midland

Courses that teach honest evaluation

The bus bay experiment needs Python, numpy and a firm grasp of training, validation and test data. These courses teach it; the trial lesson finds the level.

AI and Machine Learning for Teens course thumbnail
Ages 13 to 17

AI and Machine Learning for Teens

Machine learning with proper train, validation and test splits from the start.

See the syllabus
 Data Science Course course thumbnail
University and adult

Data Science Course

Experiments, evaluation and reporting results that hold up.

See the syllabus
 AI and ML Masterclass course thumbnail
University and adult

AI and ML Masterclass

Models from first principles to validated, deployable work.

See the syllabus
 Python and AI for Kids course thumbnail
Ages 8 to 12

Python and AI for Kids

First Python and AI projects with fair tests.

See the syllabus

Browse the course atlas for more than one hundred options and use the coding roadmap to check prerequisites.

The four we are known for

Python, AI, vibe coding and agentic coding

These run underneath everything above. Every one is live and online, placed by ability rather than by age, and the first class is free.

Vibe Coding for Teens course thumbnail
Vibe codingAges 13 to 17

Vibe Coding for Teens

Python, web and AI projects where the learner still owns the thinking.

See the syllabus
 Python Automation Course course thumbnail
Python and AICollege and adult

Python Automation Course

Automate the work you already do, then let AI carry part of it.

See the syllabus
 AI and Machine Learning for Teens course thumbnail
AI and MLAges 14 to 18

AI and Machine Learning for Teens

Train a model, read what it learned, and be able to say why it is wrong.

See the syllabus
 Codex & Claude Code course thumbnail
Agentic codingProfessionals

Codex & Claude Code

Run AI coding agents on real work without losing control of the codebase.

See the syllabus
Midland in the Census

Small households and many townhouses

Census 2021 numbers for Midland (ABS code SAL50947), paired with Western Australia's.

Midland residents and homes, 2021
MeasureMidland
Residents6,335
Median age39
Families1,366
Private dwellings3,868
People per household1.9
Midland compared with Western Australia
IndicatorMidlandWestern Australia
Semi-detached, row or townhouse homes50.6% (1,468)13.0%
Flats or apartments16.8% (487)6.5%
Separate houses32.2% (933)79.7%
Employed residents working from home3.5% (98)7.5%

Young residents

Midland counted 316 children under five, 259 aged 5 to 9 and 215 aged 10 to 14.

Older residents

Residents aged 85 and over numbered 275, 4.3% of the suburb.

Size

The suburb boundary covers about 4.2 square kilometres east of central Perth.

Figures are as published by the bureau, which nudges small cells for privacy. Adding them together would compound those nudges, so we do not.

Signature project

Choosing on the test set

Our selection: the 665 Perth Metro stop records within 6 km of Midland station, 136 of them with Y in the bay field. Each rule is a single word from the stop names: "bay if the name contains this word", or the reverse. That gives 852 rules from 426 distinct words.

Averages over 500 random half-and-half splits (our calculation)
MeasurementAccuracy
Top rule, scored on the half used to choose it81.32%
Same rule, scored on the other half79.95%
Always guessing "no bay", on the other half79.52%
How the apparent gain shrinks
Compared with always "no bay"Points gained
As reported on the reused test half1.80
On fresh data0.43

Why the score inflates

With 852 rules, some will fit any particular half by luck. Choosing the top scorer picks up that luck, and reporting its score on the same half counts the luck as skill.

How often

In 64.2% of the 500 splits the chosen rule did worse on the fresh half than on the half used to pick it.

The honest finding

Single words in Midland-area stop names barely predict bays. The model is only fractionally better than guessing, and only fresh data reveals that.

The three-way split

Keep training data for fitting, validation data for choosing between models or settings, and a test set that is opened once, at the end. Every time a test score influences a decision, that test set becomes validation data and its scores stop being trustworthy.

The Public Transport Authority of Western Australia publishes the stop file; drawing a 6 km circle around the Midland platform record was our decision. All splits, rules and accuracies are our calculation (numpy, seed 2026). The bay field describes the file, not a street survey.

Evaluating models

Keeping the test set clean

Test-set reuse rarely looks like cheating. It looks like ordinary tinkering: one more feature, one more setting, checked against the same test score each time.

Habits that protect a test set
HabitHowMidland lesson
Split three waysTraining, validation, testChoose rules on validation only
Lock the test setScore it once, at the endAvoids the 1.80-point illusion
Report a baselineAlways guess the majority class79.52% here
Repeat the splitAverage over many random splits500 splits, seed 2026

Primary learners

Practise spelling with one word list and get tested on a different one, then talk about why that is fairer.

Secondary learners

Rebuild the Midland experiment, then add a validation split and watch the reported score fall to the truth.

Adults

Ask of any dashboard model: was the reported score measured on data that influenced its design?

Public leaderboards suffer the same effect at scale: when many teams tune against one fixed test set, the top scores drift upward without real improvement. Holding back unseen data is the standard defence.

Around Midland

A rail terminus and regional coaches

What the stops file and school calendar say about Midland in particular.

Stop mix

Our boundary check finds 69 stop records in Midland: 58 standard, 6 school, 3 regional coach and 2 rail platform records.

Main roads

Great Eastern Highway and The Crescent carry 8 stop records each; Great Northern Highway and Lloyd Street 7 each.

Term 4, 2027

WA public schools hold Term 4 of 2027 from Monday 11 October to Thursday 16 December.

Midland suburb only

Census and stop figures here describe the suburb of Midland in the City of Swan. Midvale, Middle Swan, Woodbridge, Viveash, Hazelmere and Bellevue border it and are counted separately, although the 6 km project area reaches into them.

Learning ladder

From fair spelling tests to fair model tests

Ages are indicative; the free trial lesson sets each learner's rung.

Ages 6 to 10

Scratch and logic

Games and puzzles with fair rules.

Scratch Coding for KidsProblem Solving and Computational Thinking for Kids
Ages 10 to 14

Python and AI ideas

First code and simple models trained on examples.

Python and AI for KidsAI Literacy for Kids (Ages 8 to 14)
Ages 14 to 18

Machine learning

Training, validating and testing models properly.

AI and Machine Learning for TeensData Science for Teens
University and adult

Applied machine learning

Experiment design and evaluation for real work.

Data Science CourseAI and ML Masterclass
Coding and AI

AI can run a thousand experiments. Who keeps score honestly?

The person who designs the evaluation. Automated search makes test-set reuse easier, not harder.

An AI assistant asked to "find a rule that predicts bus bays" can try hundreds of candidates in seconds and report the winner's accuracy. If it scored them all on one test half, that headline figure carries the luck of the search.

Learners who have watched 81.32% fall to 79.95% on fresh data insist on a locked test set and a baseline. That is what separates a real result from an artefact of searching.

In Midland, the learner who guards the test set reports results that survive contact with new data. The longer argument is in why teenagers should still learn to code in 2026.

Delivery

Midland lessons on video

Lessons are live online, so learners join from home.

Results read aloud

Every score is read together and questioned before anyone believes it.

Following WA terms

Plans fit the state school year, with holiday breaks.

Free trial

One full lesson free, then a frank recommendation.

Fifteen to twenty learners

Groups by level, or one-to-one teaching.

Weekly pattern

A pair of fixed lessons every week, near enough eight per month.

Across time zones

Teaching happens from India, so a Midland class at 6:30pm is 4pm India time.

Stable lesson times

Since a trial ended in March 2009, Western Australia has had no daylight saving (IANA time zone database); India has none either, so times stay fixed.

Fees

Midland class fees

Three prices, the same for every learner.

First classUSD 0

Lesson one is free and finishes with a written next step.

Group tuitionUSD 100

About eight group lessons a month, fifteen to twenty in each class.

Private tuitionUSD 150

About eight private lessons a month, one teacher per learner.

Fees are charged in US dollars, and nothing is paid until after the trial when a course and timing have been chosen. See the pricing page for holiday and absence rules.

Request placementCheck availability
Reviews

Reviews posted on Google

Rated 4.9 across 547 Google reviews. These are real reviews, reproduced as written.

★★★★★

"The one step solution for my son. Modern Age Coders make learning coding so simple that kids love it. The teachers explain complex concepts clearly with practical exercises and interactive content."

Ria Mukherjee

Parent

★★★★★

"Modern Age Coders has been a game-changer for me. I struggled to grasp IT concepts and coding before joining, but their classes transformed everything. I can now confidently write complex programs with ease."

Samriddha Mondal

Student

★★★★★

"One of the most wonderful education centres out there. Education is not limited to school syllabus but focuses on skill development."

Vansh Agarwal

Student

★★★★★

"My child Dhairya is really enjoying the Modern Age Coders classes. This is his first online class and he eagerly looks forward to it. I can already see his improvement, and the teachers are very cooperative."

Sonam Oswal

Parent of Dhairya

★★★★★

"Modern Age Coders have wonderful teachers who teach in a clear, easy and practical way. The teacher boosts students' confidence and inspires them to learn without hesitation."

Sonu Goyal

Parent

★★★★★

"I highly recommend this computer coding class! The teachers are incredibly knowledgeable and passionate about coding."

Ritu Kedia

Parent

Free placement class

Book a free Midland class

Write a line about who will be learning and their age. Teenagers could rebuild the test-set experiment, and children could open with a fair-test Scratch game.

+61

FAQ

Midland AI and programming questions

Questions Midland families and adults frequently ask.

How many people live in Midland, WA?

The 2021 Census QuickStats count 6,335 residents in the suburb of Midland, with 3,868 private dwellings and 1.9 people per household.

What is test-set reuse?

Using the same test data repeatedly to choose between models or settings. Each choice leaks information from the test set, so its final score overstates how the model will do on new data.

What is a validation set?

A portion of data kept separate from training and used to compare models or tune settings, so the test set can stay untouched until the final check.

What did the Midland experiment show?

Selecting the highest-scoring of 852 one-word rules on a test half gave 81.32% accuracy on that half but 79.95% on fresh data, barely above the 79.52% of always guessing "no bay".

When can my child start AI?

Midland children can begin at six with Scratch games that show how computers learn from examples; building real models in Python comes nearer thirteen.

Are Claude courses for teenagers?

No. Only adults of 18 or over join our Claude-based courses, and teenagers learn with other AI tools.

Is there a classroom in Midland?

No. Lessons are live online only, we have no Australian office, and our listed phone number is Indian.

What do lessons cost for Midland learners?

The first lesson is free. Group classes cost USD 100 a month for two live lessons a week, about eight a month, in groups of fifteen to twenty, and one-to-one classes cost USD 150 a month. You agree the course and time before paying.

Who is in each group?

Fifteen to twenty learners at a similar level and pace, not sorted by age or suburb. One-to-one lessons suit anyone no group time fits.

Will Midland groups cover Copilot Studio?

No. Copilot Studio is one to one only; groups code their agents in Python.

Explore Perth and Western Australia

More for Midland learners

See the Perth page and coding for Perth kids. Learners heading inland will find a Kalgoorlie page, and our Western Australia guide lists every region.

Book the free classBook a priority demoMessage us on WhatsApp
Free classPriority demoCall us

Keep exploring Modern Age Coders

Related courses and guides

Learn more

Free resources

From the blog

Start here