What does an AI agents class teach Mount Waverley learners?
How to set an AI agent working on a long job and make sure it can stop and carry on without repeating itself. Mount Waverley households lean towards this kind of work: on Census day 2021, 35.6% of employed residents worked from home, well above the state.
An AI agent asked to work through every row of a big file will, sooner or later, be stopped halfway: a laptop lid closes, a network drops, a usage limit is hit. What happens next depends on one design choice made at the start. Either the job remembers how far it got, or it starts again from row one and quietly writes some results twice.
Facts last verified 11 October 2026. Teaching is online; no Mount Waverley branch is claimed.
These courses all build programs that run for a while on their own, the setting for the checkpoint project. The free first class settles the starting level.

Projects built with an AI partner, then made reliable by hand.
See the syllabus
Getting real work out of AI tools, including long multi-step tasks.
See the syllabus
Scripts and automations that save progress and recover cleanly.
See the syllabus
Coding agents on real repositories, with reviews and restarts built in.
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 Mount Waverley suburb (ABS code SAL21816) from the 2021 Census QuickStats, next to Victoria. The City of Monash page covers the wider council area.
| Headline | Mount Waverley |
|---|---|
| Residents | 35,340 |
| Median age | 40 |
| Families | 9,879 |
| Private dwellings | 13,873 |
| Average people per household | 2.7 |
| Measure | Mount Waverley | Victoria |
|---|---|---|
| Separate houses | 68.6% (8,561) | 73.4% |
| Flats or apartments | 21.5% (2,681) | 12.1% |
| Workers who drove to work | 42.6% (7,256) | 49.9% |
| Workers who took the train | 2.6% (436) | 1.6% |
| Employed residents who worked at home | 35.6% (6,055) | 25.7% |
Just over a fifth of Mount Waverley homes were flats or apartments, nearly double the Victorian share.
The bureau puts the suburb at 15.18 square kilometres, within the City of Monash.
The 2021 Census was held during a Victorian lockdown, so the work-from-home share describes that August week rather than a normal one.
QuickStats applies small random changes to each cell for confidentiality, so a column of parts may miss its own total. We print figures exactly as released and never add them up ourselves.
The state station patronage file has 1,555 station-year rows from FY18-19 to FY24-25. We wrote a scripted job that handles one row per task, the way an agent works through a list, and after every task gave it a small random chance of being cut off. Each setting ran 100 times with different random seeds. This is our simulation, not a measurement of any real agent.
| Progress saved | Chance 0.1% per task | Chance 0.2% per task |
|---|---|---|
| Never | 3,440 tasks, 1,884.9 duplicates | 12,303 tasks, 10,747.5 duplicates |
| Every 100 tasks | 1,626 tasks, 71.4 duplicates | 1,727 tasks, 171.8 duplicates |
| Every 10 tasks | 1,562 tasks, 7.5 duplicates | 1,570 tasks, 14.9 duplicates |
| After every task | 1,555 tasks, 0 duplicates | 1,555 tasks, 0 duplicates |
Doubling the interruption chance from 0.1% to 0.2% more than tripled the work for a job with no checkpoint, from 3,440 tasks to 12,303.
Saving every 10 tasks kept the extra work under 1% at both rates, because each interruption costs at most the tasks since the last save.
Duplicates do not crash anything. They sit in the output and inflate any total taken from it, which is the harder failure to spot.
Writing each result under its own key, one row per station and year, makes a repeat harmless: the second write replaces the first. With keyed writes, even a job that restarts from zero ends with exactly 1,555 rows. It still wastes the time, which is what the checkpoint saves.
The 1,555 rows come from the published patronage file; the job, the interruptions and every average above come from our own script. Real agents fail in less tidy ways, so the numbers show the shape of the problem rather than a forecast.
A checkpoint is a small file that records the last task finished. On start, the job reads it and skips ahead. The order of the two writes matters: save the result first, then move the bookmark.
| Habit | What it prevents | Cost |
|---|---|---|
| Checkpoint after each task | Redoing finished work after a stop | One small file write per task |
| Keyed or upsert writes | Duplicate rows when a task does run twice | A unique key for every result |
| Result before bookmark | A bookmark that points past a lost result | Nothing; it is only an order |
If the bookmark moves before the result is safely written, an interruption between the two silently loses a row. If the result is written first, the worst case is one task done twice, and keyed writes absorb that. Agent frameworks, data pipelines and long model training runs all use versions of the same idea.
Save points in a Scratch game: what has to be remembered so a player can carry on?
Write a Python job over the station file, pull the plug mid-run, and make it resume cleanly.
Add checkpoints and keyed writes to an agent workflow that runs overnight on work data.
From the station entry file and the 2025 school locations file.
Entries reached 658,250 in FY24-25, 60th of 222 metropolitan stations, against 820,100 in FY18-19.
Jordanville, also inside the suburb boundary, recorded 388,950 entries, down from 498,850 six years earlier.
The file places 10 open schools in the suburb, seven of them government, including Pinewood Primary School and Mount Waverley Secondary College.
Term 3, 2027
Victorian government schools run Term 3 from Monday 12 July to Friday 17 September 2027, a natural point to start or change a course.
Ages are only a guide; the free first class places each learner.
Games with levels and save points, built and debugged in blocks.
Scratch Coding for KidsVibe Coding for KidsFiles, loops and a first program that reads and writes data.
Python and AI for KidsAI Literacy for Kids (Ages 8 to 14)Building with an AI, then making the result dependable.
Vibe Coding for TeensAI Tools Course for Teens (13+)Automation and coding agents that run for hours and recover.
Python Automation CourseVibe Coding for CollegeThe person who asked for the loop, because the agent will rarely add a checkpoint unless told to.
Ask an AI to process a file row by row and it will usually produce a clean, correct loop that works perfectly when nothing goes wrong. It is the first interruption that exposes the missing bookmark, and by then the output may already hold thousands of repeated rows.
A learner who has watched 1,555 tasks swell to 12,303 knows to ask for a checkpoint, keyed writes and the right order of saves before the job ever starts. That is the difference between vibe coding something and owning it.
The Mount Waverley learner who plans for interruption is the one whose agent can be trusted to run unattended. The longer argument is in why teenagers should still learn to code in 2026.
Classes take place over video, so learners join from home with nothing to travel to.
Learners write on their own machine with the screen shared, and the teacher steps in as the code grows.
Courses run in Victorian school terms and pause for the holidays.
The first class is free and complete. At the end the teacher recommends a course and a weekly time.
Groups hold fifteen to twenty learners who are at a similar point.
Two lessons on fixed days each week, around eight a month, with holidays known ahead.
Teachers are based in India, so a 4:15pm Mount Waverley lesson in daylight saving is 10:45am India time, or 11:45am in Melbourne winter.
Melbourne clocks
Daylight saving in Melbourne began on Sunday 4 October 2026 and ends on Sunday 4 April 2027. India keeps one time all year, so lesson times are always confirmed in Melbourne time.
Three prices cover every course.
A free opening lesson that ends with a clear next step.
A month of group lessons, generally eight, in a group of fifteen to twenty at one level.
Roughly eight one-to-one lessons a month with the same teacher.
We bill families outside India in US dollars, so Mount Waverley pays exactly what other overseas families pay. No payment is taken until the trial has fixed the course and the time; the pricing page sets out holidays, missed lessons and changes of format.
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
Share an age and an interest. An eight-year-old might start with a save-point game in Scratch; a Year 10 student might start by building an agent job that can be stopped and resumed.
Questions from Mount Waverley parents, students and working adults.
The 2021 Census QuickStats for the Mount Waverley suburb record 35,340 residents in 9,879 families, with 13,873 private dwellings and a median age of 40.
A saved record of how far the job has got, usually the last task completed. When the job restarts, it reads the checkpoint and carries on from there instead of starting over.
Building software by describing what you want to an AI and shaping its code through conversation. Our classes add the part that makes it safe: reading, testing and fixing what the AI wrote.
With a 0.2% chance of interruption per task, our simulated job ran 12,303 tasks on average to finish 1,555 when it never saved progress, and exactly 1,555 when it saved after every task.
Yes, with sensible tools and supervision. Teenagers learn to plan an agent's steps, check its output and make it recover from failure, which are habits worth having at any age.
Children from six start with Scratch, building games with levels, scores and save points.
Online only. We have no Australian premises, and the phone number on our site is an Indian one.
The first lesson is free. After it, group classes cost USD 100 a month for two live lessons a week, about eight a month, with fifteen to twenty learners, and one-to-one classes cost USD 150 a month. Course, format and time are agreed before any payment.
Fifteen to twenty learners grouped on level, pace and goals, or one to one if no group time fits.
Many do. Working adults join their own groups or learn one to one, often to automate a slow task at work with an agent they can actually rely on.
The City of Monash page covers the council area around Mount Waverley, and the Glen Waverley page the suburb next door. For parents of younger children there is a guide to kids coding classes in Melbourne, and the Victoria guide covers the state.