What are the best vibe coding and AI agents classes in Bunbury?
Bunbury teenagers and adults learn to build software alongside an AI assistant, then to check and correct its work until their agents behave as intended. Bunbury's urban area counted 76,452 people at the 2021 Census, and 18.4% of employed residents worked in trades and technical jobs. Lessons run online, one to one or in small level-matched groups, and the first is free.
People type place names badly. An AI agent that answers "which schools are near Austrlind?" needs a lookup tool that forgives the missing letter and still refuses to invent a match for a town it has never heard of. Our Bunbury project builds that tool from the Western Australian Department of Education's open list of schools: 44 schools within 25 kilometres of central Bunbury, in 16 suburbs. We wrote the Jaro-Winkler similarity score from its definition, generated 192 deliberate typos and tested. Every typo found its right suburb. The hard part turned out to be elsewhere: the weakest genuine typo scored 0.8905, and an unrelated suburb from elsewhere in the state scored 0.89.
Facts last verified 7 October 2026. Teaching is online; no Bunbury branch is claimed.
All four connect to the suburb-matching project. The free lesson chooses the right one, and booking requires no card.

Python and web projects built with an AI assistant, then read, tested and repaired line by line.
See the syllabus
Strings, functions and tests, the Python that every agent tool is made from.
See the syllabus
Language models, tool calling and agents, designed and measured rather than taken on trust.
See the syllabus
Full applications built alongside AI, with the data structures and testing that keep them correct.
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 syllabusThe bureau's QuickStats for the Bunbury Significant Urban Area, set against Western Australia and Australia.
| Measure | Bunbury | Western Australia | Australia |
|---|---|---|---|
| Population | 76,452 | no figure here | no figure here |
| Median age | 39 | no figure here | no figure here |
| Families | 20,598 | no figure here | no figure here |
| Technicians and trades workers, share of employed | 6,573 (18.4%) | 15.3% | 12.9% |
| Separate houses, share of occupied homes | 24,294 (85.5%) | 79.7% | 72.3% |
| Teenagers aged 15 to 19 | 4,749 (6.2%) | 5.8% | 5.7% |
6,573 employed residents worked as technicians or trades workers, a larger share than in Western Australia overall.
The urban area had 32,807 private dwellings and 2.5 people per household; only 610 occupied homes were flats or apartments.
A bachelor degree or above was held by 8,911 residents aged 15 or older, 14.5%, while 5,826 of those in work were classed as professionals.
Census counts are nudged slightly at random by the bureau to protect privacy, so the parts of a table do not always add up to its total. We use each published figure as it stands and never produce our own totals by summing parts.
A place-name lookup tool for an AI agent, built on the Department of Education's open schools list and scored with Jaro-Winkler similarity.
| Item | Figure |
|---|---|
| Schools in the statewide list | 1,152 |
| Schools within 25 km of central Bunbury | 44 |
| Distinct suburb names among them | 16 |
| Deliberate typos generated: swapped, dropped and doubled letters | 192 |
| Suburb names elsewhere in the list, used as impostors | 483 |
| Accept a match scoring at least | Typos correctly accepted | Impostors wrongly accepted |
|---|---|---|
| 0.85 | 192 of 192 | 6 of 483 |
| 0.90 | 191 of 192 | 0 of 483 |
| 0.95 | 182 of 192 | 0 of 483 |
Jaro-Winkler compares two words by their shared letters, how far those letters moved and whether the first few match. Identical names score 1.
For each query, find the closest of the 16 local suburbs. On all 192 typos the closest was the right one, for Jaro-Winkler and for edit distance alike.
Ranking always returns something. The real design question is the cut-off below which the tool says "no match" instead.
The weakest genuine typo, BUUREKUP for BUREKUP, scored 0.8905. The strongest impostor, SOUTH YUNDERUP from elsewhere in the state, scored 0.89 against SOUTH BUNBURY because the two share a long opening. There is almost no clear water between them. A threshold of 0.90 rejects every impostor and loses one real typo; 0.85 keeps every typo and admits six strangers. Choosing between those is a product decision, and an agent builder has to make it deliberately.
All scores are ours: Jaro-Winkler written from its definition, checked against the standard textbook examples, and run on the department's list as downloaded on 7 October 2026 with typos generated by a fixed random seed. A query for BUSSELTON still returns BUREKUP as its closest local name, at 0.67, which is exactly why a threshold is needed. We use school locations only to define the area; no school information beyond suburb names appears in the project.
An agent cannot see inside its tools. Everything it needs to behave well has to come back in the result.
| Situation | Example | What the tool should return |
|---|---|---|
| Exact name | EATON | The match, with a score of 1 |
| Small typo | BUNBRY | BUNBURY, scored 0.971, flagged as corrected |
| Missing letter | AUSTRLIND | AUSTRALIND, scored 0.98, flagged as corrected |
| Unknown town | PERTH | No match: the closest, EATON, scores only 0.60 |
| Near neighbours | SOUTH BUNBURY and EAST BUNBURY | Kept apart: they score 0.7675 against each other |
Vibe coding gets the first version of this tool written in moments: describe it, and an AI produces a fuzzy matcher. Our learners then do the engineering. They build the typo set, find the impostors, plot where the scores of each fall and pick a threshold with their eyes open. The tool returns the score and a "corrected" flag, so the agent can tell the user it assumed AUSTRALIND rather than silently changing the question.
Write Jaro-Winkler from its definition in Python, test it on famous examples, then on Bunbury suburb names.
Compare similarity measures, build evaluation sets with impostors and choose thresholds with precision and recall.
Clean messy names in work data and build lookups that say when they are unsure.
We have no tie to the WA Department of Education, the state government, the City of Bunbury or the statistics bureau, and do not suggest one. The bureau supplied the population figures; our scripts produced every distance, score and threshold on this page.
Two calendar facts that matter for a Bunbury learner's week.
Runs from Monday 1 February to Friday 9 April, according to the WA Department of Education's published dates.
Runs from Monday 11 October to Thursday 16 December, finishing on a Thursday.
Western Australia keeps Western Standard Time all year, so lesson times never shift with the seasons.
Names are data too
Suburb names in the department's list are stored in capitals, with spaces inside some, such as GLEN IRIS and BRUNSWICK JUNCTION. A matcher that ignores spaces or case differently from the data will score names inconsistently, so learners normalise both sides the same way before comparing anything.
Ages give a rough idea; the free lesson decides where each learner starts.
Python strings, loops and functions that compare names letter by letter and count the differences.
Python for TeensVibe Coding for TeensJaro-Winkler and edit distance, built by hand and tested on Bunbury typos and impostors.
Vibe Coding for TeensAI Tools Course for Teens (13+)Tool interfaces, evaluation sets and thresholds for agents that call real data.
Generative AI CourseVibe Coding for CollegeCoding agents and automation for professional tasks, with tests that keep them honest.
Codex & Claude CodePython Automation CourseBecause without a threshold the agent will happily turn BUSSELTON into BUREKUP and answer a question nobody asked.
A fuzzy lookup always has a closest match. Ask about a town outside the list and the tool still returns its nearest neighbour, and an agent built on it will pass that answer on confidently. Our Bunbury tests show how thin the line can be: a real typo at 0.8905 and an impostor at 0.89. Where the cut-off sits decides whether the agent occasionally refuses a fair question or occasionally answers the wrong one.
That decision is invisible to anyone who only vibe codes the matcher. Learners who have built the test set see it immediately and design around it: return the score, flag corrections, refuse below the threshold, and tell the user what was assumed. Those are the habits that make an agent safe to put in front of other people.
In 2026 any AI can produce a fuzzy matcher; a Bunbury learner who can show where it is safe to trust is the one others will want building their agents. The longer argument is in why teenagers should still learn to code in 2026.
Greater Bunbury spreads around Koombana Bay and the estuary, and a video lesson reaches every suburb on it.
A learner in Carey Park and another in Gelorup can work in the same class without either travelling.
Teachers follow WA year levels and the state's four terms, and senior students can relate projects to their own courses. Teaching is in English.
Your free first session is a proper lesson that ends with a level, a course suggestion and a time. No card is ever requested.
Fifteen to twenty learners at the same stage, from Bunbury and elsewhere, so regional learners always find classmates.
Around eight a month in a fixed slot, with school holidays and exam periods planned with the teacher.
Western Australia and India both skip daylight saving, so a 6pm lesson in Bunbury is 3:30pm for our teachers all year round.
The smallest gap in Australia
Bunbury runs on Western Standard Time, just two and a half hours ahead of India, the shortest difference of any Australian state. Lesson times are always confirmed in Bunbury time.
Three straightforward prices.
A full first lesson for free, after which you get a level, a suggested course and a time.
A month of lessons, usually eight, in a group of fifteen to twenty learners at your level.
About eight lessons a month with your own teacher, one to one.
Outside India all our fees are set in US dollars, which means Withers and College Grove households receive matching invoices and WA has no price list of its own. Payment begins once the trial has pinned down a course and a lesson slot, and our pricing page answers questions about breaks, absences and changing 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. A Year 9 student might begin by comparing two misspelt names letter by letter; an adult might build the whole lookup tool and its tests.
Answers to what Bunbury learners and parents raise most.
The 2021 Census counted 76,452 people in the Bunbury urban area, according to the statistics bureau's QuickStats. The median age was 39.
A score between 0 and 1 for how alike two short strings are, such as names. It counts matching letters, penalises letters that appear out of order and gives extra weight to a shared beginning, which suits typing mistakes.
Learners take the WA Department of Education's open schools list, collect the 16 suburb names within 25 km of central Bunbury, generate 192 typos and test a Jaro-Winkler lookup. Every typo finds its suburb; the challenge is choosing a threshold that rejects names from elsewhere.
Building software by explaining the goal to an AI assistant in ordinary words and letting it draft the code. Our lessons add the part that matters: reading that draft, testing it and repairing it.
Yes. We follow WA year levels and help senior students with computing and maths. Anything submitted for school assessment is the student's own work.
Usually after school or at weekends, agreed at the trial and quoted in Bunbury time. Neither WA nor India changes its clocks, so the two and a half hour gap to our teachers never moves.
Yes. From age six children use Scratch and simple AI activities, moving to Python later. This page is written for teenagers and adults.
No, and none is claimed. All teaching is live and online, with no premises anywhere in Australia. A learner needs a laptop or desktop, working sound and a reliable connection, and the contact number shown is an Indian one.
The first lesson is free. After that a group place is USD 100 a month for two live lessons a week, around eight a month, with fifteen to twenty learners, or USD 150 a month for the same pattern one to one. Course, format and time are agreed before anything is paid.
Fifteen to twenty learners, matched on level, pace and goals rather than on age or suburb. If no suitable group meets at a workable time, one-to-one lessons are offered instead.
The Western Australia page covers the whole state, and Perth has its own Perth Python page. For more on keeping AI-written code honest, read understand the code, don't copy-paste. The Australia coding page lists every page we have written for Australian learners.