What do Eight Mile Plains learners gain from AI and programming classes?
The ability to build machine learning models and the discipline to test whether a recommended technique actually helps on their own data. Children begin with Scratch at six; teenagers and adults train models in Python. Eight Mile Plains recorded 15,326 residents at the 2021 Census.
Textbooks say k-means++ picks better starting centres than random guesses. We tried both, a thousand times each, on the 47 bus stop records inside Eight Mile Plains. The fancy method won on some measures and lost on another, which is exactly why learners should run the experiment rather than quote the rule.
Facts last verified 12 October 2026. Teaching is online; no Eight Mile Plains branch is claimed.
The stop-clustering project needs Python, numpy and the habit of comparing methods fairly. These courses build each part; the trial lesson finds the level.

Clustering, classification and regression built by hand, then compared honestly.
See the syllabus
Machine learning from first principles to evaluated, working models.
See the syllabus
Python data work with maps, charts and repeatable experiments.
See the syllabus
First Python and AI projects, small enough to test by eye.
See the syllabusBrowse the course atlas for more than one hundred options and use the coding roadmap to check prerequisites.
The four we are known for
These run underneath everything above. Every one is live and online, placed by ability rather than by age, and the first class is free.

Python, web and AI projects where the learner still owns the thinking.
See the syllabus
Automate the work you already do, then let AI carry part of it.
See the syllabus
Train a model, read what it learned, and be able to say why it is wrong.
See the syllabus
Run AI coding agents on real work without losing control of the codebase.
See the syllabusA selection of 2021 Census QuickStats for the suburb of Eight Mile Plains (ABS code SAL30950), shown beside the Queensland figure.
| Measure | Eight Mile Plains |
|---|---|
| Residents | 15,326 |
| Median age | 36 |
| Families | 4,125 |
| Private dwellings | 5,631 |
| People per household | 2.8 |
| Indicator | Eight Mile Plains | Queensland |
|---|---|---|
| Workers who rode a bus to work | 7.2% (534) | 1.7% |
| Semi-detached, row or townhouse homes | 26.1% (1,348) | 11.7% |
| Employed people in professional jobs | 29.5% (2,184) | 21.4% |
| Adults with a bachelor degree or higher | 38.9% (4,920) | 21.9% |
| People in education who attend university | 22.0% (1,045) | 14.2% |
The suburb had 947 residents aged 5 to 9 and 843 aged 10 to 14.
Residents aged 85 and over numbered 362, so lessons here span grandchildren and grandparents.
The ABS boundary takes in roughly 7.6 square kilometres of Brisbane's southern suburbs.
The bureau adds small privacy perturbations to Census cells, so these figures are reproduced exactly and never summed into totals of our own making.
k-means groups points around k centres, but it needs starting centres, and a bad start can trap it in a poor answer. Random seeding picks k stops at random. k-means++ picks one at random, then favours stops far from the centres already chosen. We ran each method 1,000 times on the 47 stop records with k = 4, 5 and 6.
| Measure | Random start | k-means++ |
|---|---|---|
| Median typical distance from stop to its centre | 494 m | 493 m |
| Worst run | 632 m | 596 m |
| Runs reaching the tightest answer found (470 m) | 10.5% | 9.3% |
| Average iterations to settle | 5.22 | 4.43 |
| k | Random: median, worst | k-means++: median, worst |
|---|---|---|
| 4 | 574 m, 676 m | 571 m, 666 m |
| 6 | 450 m, 591 m | 440 m, 558 m |
Its worst runs were tighter at every k, and it settled in fewer rounds, about 4.4 against 5.2 at k = 5.
Medians were within 10 metres. At k = 5, random seeding actually reached the tightest clustering more often, 10.5% of runs against 9.3%.
Neither method found the tightest answer reliably in one go. Running several restarts and keeping the tightest is what fixes a bad start.
With only 47 stops in one suburb, almost any start lands near a good answer, so the plain baseline holds its own. k-means++ was designed for large datasets where random starts often fail badly. Reporting the tie is the result; claiming a win would be repeating a textbook, not measuring Eight Mile Plains.
Stop positions come from the Translink timetable feed; membership of the suburb is our point-in-boundary check. Distances use a flat projection in metres, fine at suburb scale. The typical distance is the root mean square distance from each stop to its cluster centre. All runs, seeds and statistics are our calculation with numpy, seed 2026.
The interesting part of this project is not k-means itself but the experiment around it.
| Choice | Why | Here |
|---|---|---|
| Many runs per method | One run is luck, not evidence | 1,000 each |
| Same data and seed source | Differences must come from the method | 47 stops, seed 2026 |
| Several values of k | A result at one k may not hold at another | k = 4, 5 and 6 |
| Report medians and worst cases | Averages hide the bad runs | Worst 632 m against 596 m |
Drop counters on a map, pick meeting points, and see how the first guess changes the final groups.
Code k-means in numpy, add k-means++ seeding, and run both enough times to make a fair call.
Question any "standard practice" in an analytics tool by testing it on your own data first.
Library defaults are worth reading. In scikit-learn, k-means++ is the default seeding and, with the default n_init setting, it runs from a single start, while random seeding runs 10 times and keeps the tightest result. Our runs suggest restarts deserve as much thought as seeding.
Local markers from the timetable feed, the council park register and the Queensland school calendar.
On Wednesday 21 October 2026, the 47 stop records inside the suburb had 2,195 scheduled bus visits on 9 routes, by our count.
Warrigal Rd at Fanfare, at Bordeaux and at Padstow Views tied that day with 139 scheduled visits each.
Queensland state schools open on Wednesday 27 January 2027, and Term 1 ends on Thursday 25 March.
Just this suburb
Brisbane City Council's park register lists 35 parks in Eight Mile Plains. Kuraby, Macgregor, Rochedale, Runcorn, Sunnybank, Upper Mount Gravatt and Wishart all border the suburb, and their figures stay separate from these.
Ages are a guide; the trial lesson shows where to begin.
Puzzles and games that build step-by-step thinking.
Scratch Coding for KidsProblem Solving and Computational Thinking for KidsFirst code and models that learn from examples.
Python and AI for KidsAI Literacy for Kids (Ages 8 to 14)Clustering, regression and fair evaluation in Python.
AI and Machine Learning for TeensData Science for TeensData science and models for study and work.
Data Science CourseAI and ML MasterclassThe learner. AI assistants repeat standard advice faithfully, but that advice was measured on someone else's data.
Ask an assistant how to cluster bus stops and it will recommend k-means++ with confidence. Good advice in general. On the Eight Mile Plains stops, though, it made almost no difference to the typical result, and random seeding found the tightest clustering slightly more often.
Learners who have run the comparison themselves know to treat advice as a hypothesis. They set up a fair test, report what it shows, and change course when the data disagrees. That habit is what AI cannot supply.
In Eight Mile Plains, the learner who tests the advice is the one ready to use AI well. The longer argument is in why teenagers should still learn to code in 2026.
Lessons happen live over video, so learners anywhere in the suburb join from home.
Teacher and learner run code and read the results together, line by line.
Plans follow Queensland terms, with breaks in the holidays.
One real lesson at no charge, closing with an honest recommendation.
Fifteen to twenty learners of similar level, or one to one.
Set weekly times, close to eight lessons a month.
Our teachers are based in India, so a 7pm class here starts at 2:30pm India time, all year.
Clocks that never change
Brisbane has had no daylight saving since March 1992 (IANA time zone database), and India has none either, so agreed lesson times hold year round.
Three prices for every family.
The trial lesson is free and ends with a written plan.
About eight group lessons a month with fifteen to twenty learners at the same level.
About eight private lessons a month, one teacher to one learner.
Payment is in US dollars and comes after the trial, once the course and time are agreed. Holiday breaks, missed lessons and format changes are covered on the pricing page.
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
Tell us the learner's age and interests. A teenager might start by clustering the suburb's bus stops; a younger child might start with Scratch logic puzzles.
Questions families and adult learners in Eight Mile Plains often ask.
The 2021 Census QuickStats count 15,326 residents in the suburb of Eight Mile Plains, with 4,125 families and a median age of 36.
An algorithm that splits points into k groups by repeatedly assigning each point to its nearest centre and moving each centre to the middle of its group, until nothing changes.
A way to choose starting centres for k-means. It picks the first at random and each later one with probability that grows with distance from the centres already chosen, which spreads the starts out.
Only partly. Over 1,000 runs at k = 5, medians were 493 m and 494 m; k-means++ had a better worst run and fewer iterations, but random starts found the tightest clustering more often.
From six with Scratch and simple AI activities. Training models in Python usually starts around thirteen.
No. Our Claude-based courses are for people aged 18 and over only; teenagers use other AI tools.
No. Lessons are live online only, we have no Australian office, and our listed phone number is Indian.
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.
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.
No. Copilot Studio is taught one to one only; group learners build agents in Python code.
Try the Brisbane guide or coding for Brisbane children. Families to the south may also want Logan or the Gold Coast, and the Queensland guide covers the rest.