What do AI and programming classes look like for Berwick learners?
Live online lessons where learners build models and then measure how good they are, honestly. Berwick is a family suburb: the 2021 Census counted 50,298 residents in 13,846 families, with 7.7% aged 10 to 14 against 6.0% across Victoria, and 89.0% of homes were separate houses.
Berwick station had 1,008,200 entries in FY24-25, up from 864,800 in FY18-19. That rise put it eighth of 219 Melbourne stations ranked by growth. Could a model have picked the top ten in advance? More importantly, how would you know whether it did well? That second question is where most machine learning projects go wrong.
Facts last verified 10 October 2026. Teaching is online; no Berwick branch is claimed.
Each course here teaches learners to test a model before trusting it, the habit behind the ranking project. The free first class decides the starting point.

Build models in Python, then score them against simple baselines.
See the syllabus
The Python needed to sort, rank and compare real data.
See the syllabus
Machine learning end to end, including evaluation metrics for ranking.
See the syllabus
Statistics and data work that make a metric mean something.
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 syllabusFigures for the Berwick suburb (ABS code SAL20224, listed as Berwick (Vic.)) from the 2021 Census QuickStats, set against Victoria.
| Headline | Berwick |
|---|---|
| Residents | 50,298 |
| Median age | 38 |
| Families | 13,846 |
| Private dwellings | 17,981 |
| Average people per household | 2.9 |
| Measure | Berwick | Victoria |
|---|---|---|
| Separate houses | 89.0% (14,827) | 73.4% |
| Flats or apartments | 0.3% (52) | 12.1% |
| Residents aged 5 to 9 | 7.5% (3,750) | 6.2% |
| Residents aged 10 to 14 | 7.7% (3,890) | 6.0% |
| Workers who drove to work | 54.9% (13,925) | 49.9% |
Berwick homes averaged 2.1 motor vehicles each on Census night, with 2.9 people per household.
The bureau measures the suburb at 33.20 square kilometres, in the City of Casey.
Because the Census fell in the August 2021 lockdown, the 6,401 residents who worked from home, 25.2%, describe that week only.
QuickStats adjusts small counts to protect privacy, so table rows may not sum to totals. Every figure here is quoted as published; we never add parts to make a new total.
We took the 219 stations with figures in FY18-19, FY23-24 and FY24-25 and ranked them by FY24-25 entries as a share of FY18-19. Then we tried three ways of predicting that ranking and scored each one with two standard ranking metrics.
| Model | Precision at 10 | NDCG at 10 | Berwick placed |
|---|---|---|---|
| Last year's growth, FY23-24 against FY18-19 | 0.8 | 0.9898 | 8th |
| Distance from Flinders Street | 0.2 | 0.6049 | 14th |
| Weekend share of trips in FY18-19 | 0.1 | 0.487 | 210th |
| Random order, average of 1,000 runs | 0.0475 | 0.553 | Anywhere |
Assuming last year's growth continues found eight of the true top ten and put Berwick exactly where it finished, eighth.
Distance from the city and weekend share sound like sensible clues. Distance barely beat chance; weekend share did worse than chance on NDCG.
Merinda Park grew most, to 1.882 times its FY18-19 entries, then Officer at 1.697; Berwick's ratio was 1.166.
Random guessing scored 0.553 on NDCG at 10, because every station grew or shrank by a modest amount and so even a poor top ten earns partial credit. On its own, 0.6049 sounds respectable. Next to random, it shows distance from the city added very little. Always score a dumb baseline first.
The models, scores and rankings are ours, built on the department's published estimates, which are rounded to the nearest 50. They show what happened in our test; we make no claim about why any station grew.
A ranking model is judged on its top results, because those are the ones anyone reads.
| Metric | What it measures | Blind spot |
|---|---|---|
| Precision at 10 | How many of the model's top ten are in the true top ten | Ignores order inside the ten |
| NDCG at 10 | Rewards high true values placed near the top, discounted by position | Graded gains can flatter weak models |
| Spearman correlation | Agreement across the whole ranking | Treats rank 200 as seriously as rank 2 |
Last year's growth also had a Spearman correlation of 0.86 across all 219 stations, while distance from the city came out at minus 0.0429, essentially no relationship. NDCG divides the score of the model's list by the score of the perfect list, so 1 means the order was ideal.
Guess the top five in a class vote, then count how many guesses made the real top five.
Code precision at k and NDCG in Python and test them on the station files.
Score a search or recommendation feature at work against a random baseline before reporting it.
More detail from the state files.
The only metropolitan station inside the suburb, on the Pakenham line, with 1,008,200 entries in FY24-25.
The 2025 school locations file places 12 open schools in Berwick: nine government, two Catholic and one independent, including Berwick Primary School and Berwick Secondary College.
Victorian government school students start on Thursday 28 January 2027; Term 1 ends on Thursday 25 March.
Which Berwick
This page is about the Berwick suburb in Melbourne's south-east, in the City of Casey. Places with similar names elsewhere have their own figures, and none of them are used here.
Ages are a rough guide; the free first class sets each learner's level.
Games with scores, leaderboards and simple sorting.
Scratch Coding for KidsAI Literacy for Kids (Ages 8 to 14)Lists, sorting and first guesses about what makes a model good.
Python and AI for KidsMaths Through CodingTrain models, then measure them against baselines.
AI and Machine Learning for TeensData Science for TeensEvaluation, ranking and models that hold up in use.
AI and ML MasterclassData Science CourseBecause building a model takes minutes now, and knowing whether it is any good is the part that still takes a person.
Ask an assistant for a model to predict station growth and it will give you one, often with an impressive sounding feature like distance from the city. It will rarely volunteer that a random guess scores 0.553 on the same metric, or that last year's numbers beat its idea easily.
A learner who has computed precision at 10 by hand and watched a sensible feature lose to chance knows to ask for the baseline every time. That question saves more projects than any new algorithm.
The Berwick learner who measures a model against a baseline is the one who can tell real progress from a good story. The longer argument is in why teenagers should still learn to code in 2026.
Everything happens on screen, which suits a suburb where most trips mean the car.
At lesson time the learner opens the link, shares a screen and codes with the teacher watching.
Courses follow Victoria's school terms and pause in the state holidays.
The free class is a normal lesson; afterwards we recommend a course and a weekly slot.
Fifteen to twenty learners per group, matched on level and pace.
Two fixed lessons a week, about eight each month, with breaks agreed early.
Our teachers are in India: a 5:30pm Berwick lesson in summer is noon India time, then 1pm after Melbourne's clocks go back.
Daylight saving
Melbourne's daylight saving began on Sunday 4 October 2026 and ends on Sunday 4 April 2027. India has no daylight saving, so we always confirm times in Melbourne time.
Three prices, set out plainly.
The first lesson is free and finishes with a clear recommendation.
A month of group lessons, usually eight, with fifteen to twenty learners at the same level.
About eight lessons a month, one teacher with one learner.
Families outside India are billed in US dollars, so Berwick pays the same as other families abroad. Billing starts only after the trial has settled course and time; holidays, absences and changing format 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 an interest. A seven-year-old might start with a leaderboard game in Scratch; a Year 11 student might start by scoring the station models in Python.
What Berwick parents and adult learners ask before starting.
The 2021 Census QuickStats for the Berwick suburb record 50,298 residents, 13,846 families and 17,981 private dwellings.
The share of a model's top ten that are truly in the top ten. Our strongest station model scored 0.8, meaning eight of its ten picks were right.
Normalised discounted cumulative gain: a score from 0 to 1 that rewards putting the most relevant items at the very top of a list.
That the simplest model, assuming last year's growth continues, beat two cleverer-sounding ones, and that a metric only means something next to a random baseline.
From six, with Scratch and simple AI activities. Training and testing real models usually starts in the early teens, once Python is familiar.
Some helps, and we teach it as needed. Ranking metrics like these need only fractions, sorting and logarithms explained with examples.
No. All teaching is live online, we have no premises in Australia, and the phone number shown is an Indian number.
The first lesson is free. Group classes then cost USD 100 a month for two live lessons a week, about eight a month, with fifteen to twenty learners, or USD 150 a month for one-to-one lessons. Course, format and time are agreed before paying.
Fifteen to twenty learners, grouped by level, pace and goals. Private lessons are offered when no group time works.
Yes. Adults have their own groups or study privately, often to evaluate AI tools they already use at work.
For younger learners, read about coding classes for kids in Melbourne. Families weighing selective entry can see Victorian selective entry maths practice, and the Victoria guide covers the state.