What are the best vibe coding and AI agents classes for Liverpool learners?
Classes taught live over video, in which a teacher shows Liverpool teenagers and adults how to describe a program to an AI, test what it writes, and build small agents that pick the right tool for each question. Liverpool is one of Sydney's youngest council areas: the 2021 Census recorded a median age of 34 across its 233,446 residents.
Type "how many government schools are in Liverpool?" into an AI agent that can search the education department's school list. The honest answer for the council area is 49. An agent that searches school names returns 5. One that matches the suburb field exactly returns 2. One that looks for the word Liverpool in the council column returns 59, because the separate Liverpool Plains council area matches too. Same data, same question, four answers: the difference is which tool the agent chose and how it called it.
Facts last verified 10 October 2026. Teaching is online; no Liverpool branch is claimed.
Each course connects with the search project on this page. A free trial lesson decides the fit, and no card details are needed.

Describe a program in plain words, let an AI draft it, then test and fix the Python yourself.
See the syllabus
How AI tools search, summarise and slip up, and how to check their answers against the source.
See the syllabus
Full applications built with AI assistance, plus the data structures that keep queries correct.
See the syllabus
Language models, retrieval and agents, including how tool calls are designed and tested.
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 syllabusFrom the statistics bureau's 2021 QuickStats for the Liverpool local government area, alongside New South Wales and Australia.
| Measure | Value |
|---|---|
| People counted | 233,446 |
| Families | 59,471 |
| Private dwellings | 77,824 |
| People per household, on average | 3.2 |
| Area on the ABS boundary | About 306 square kilometres |
| Indicator | Liverpool | NSW | Australia |
|---|---|---|---|
| Children aged 5 to 9, share of residents | 7.6% (17,670) | 6.2% | 6.2% |
| Couple families with children, share of families | 55.8% (33,171) | 44.7% | 43.7% |
| Occupied homes that are separate houses | 73.1% (50,658) | 65.6% | 72.3% |
| Median age | 34 | 39 | 38 |
Every five-year band from birth to 14 sits above the state share: 16,664 children under five, 17,670 aged 5 to 9 and 17,236 aged 10 to 14.
Households average 3.2 people, against 2.6 across New South Wales, and more than half of all families are couples with children.
Only 16.3% of occupied homes are flats or apartments, below the state figure of 21.7%, in a council area that stretches about 306 square kilometres.
Small random adjustments protect privacy in every Census table, so parts seldom add exactly to totals. Each figure here is printed as the bureau released it; none is a sum we made ourselves.
We gave the same question to four versions of a simple agent tool, each searching the education department's 2,210-row school list a different way, and counted what came back.
| Tool call the agent makes | Schools returned | Inside the Liverpool council area |
|---|---|---|
| School name contains "liverpool" | 5 | All 5, but 44 schools missed |
| Suburb field equals "Liverpool" exactly | 2 | Both, but 47 missed |
| Suburb field equals "liverpool" after trimming and lower-casing | 7 | All 7, still 42 missed |
| Council field contains "liverpool" | 59 | 49, plus 10 from Liverpool Plains |
| Council field equals "Liverpool" | 49 | All 49, none extra |
| Tool call | Precision | Recall |
|---|---|---|
| Name contains "liverpool" | 5 of 5, 100% | 5 of 49, 10.2% |
| Suburb field, cleaned | 7 of 7, 100% | 7 of 49, 14.3% |
| Council field contains "liverpool" | 49 of 59, 83.1% | 49 of 49, 100% |
| Council field equals "Liverpool" | 49 of 49, 100% | 49 of 49, 100% |
Fast and precise, but almost useless here: most schools in the council area are named after other suburbs, from Casula to Green Valley.
Spelling decides everything. The file stores the same suburb as "Liverpool", "Liverpool " with a trailing space and "LIVERPOOL", and an exact match finds only 2 schools.
The right column for the question, but a contains-search lets in Liverpool Plains. An exact match on the field returns all 49 and nothing else.
Before calling a tool, it decides what kind of place the question means: a council area, a suburb or a name. Then it filters the field built for that meaning, uses an exact match, and reports which field it used. That turns a guess into an answer a person can check, and on this file it is the only route to 49.
Across the whole file, 1,357 of the 2,210 suburb values end in a space and 432 are written in capitals, so clean-up is not optional. Every count in this section is our own, from the department's published file; the department plays no part in the analysis, and nothing here describes any school's size or results.
Learners usually start by asking an AI to write the search tool, then improve it step by step.
| Version | Change | Effect on the count |
|---|---|---|
| First AI draft | Searches school names for the word | Returns 5 |
| Second | Searches the suburb field instead | Returns 2 until spaces and capitals are cleaned, then 7 |
| Third | Searches the council field with contains | Returns 59, picking up a different council |
| Fourth | Exact match on the council field, with the field named in the reply | Returns 49 and says how |
Each step teaches something that applies far beyond school lists: read the columns before you query, clean text before you compare it, and never let an agent hide which filter it used. The same lessons apply when an agent searches customer records, product catalogues or a company wiki.
Write the four versions in Python, print what each returns, and compare them with the hand-checked answer.
Design a tool schema that forces the agent to name the field and match type, and test it with tricky questions.
Apply the same checks to AI tools at work before trusting a count, a total or a list they produce.
The education department, Transport for NSW and the statistics bureau publish the files used here and have no link with us. All precision and recall figures are our own calculations.
More detail from the same public sources.
Luddenham Public School records its first teacher in 1860 and Liverpool Public School in 1863, while Gulyangarri Public School, with a first teacher in 2024, is the newest school in the file for this council area.
Liverpool, Casula, Warwick Farm, Holsworthy and Edmondson Park stations all sit inside the council area on our map check. Liverpool Station logged 210,605 Opal entries in August 2026.
Eastern division students start Term 1 on Wednesday 3 February 2027 and finish Term 4 on Monday 20 December 2027.
A cleaning exercise for later
The 49 Liverpool rows carry 37 different suburb labels, which shrink to 28 once spaces and capitals are normalised. Counting schools per suburb before and after cleaning is a short task that shows why data preparation comes first.
Ages are only a guide; the free trial lesson sets the real starting point.
Scratch games and simple apps made with an AI helper, while learning to spot when its suggestion is wrong.
Vibe Coding for KidsScratch Coding for KidsLoops, lists and files in Python, with AI drafts that the learner tests line by line.
Python for TeensVibe Coding for TeensSmall agents that call search tools, plus the checks that keep their answers honest.
AI Tools Course for Teens (13+)AI and Machine Learning for TeensRetrieval, tool design and evaluation for agents that handle data people rely on.
Vibe Coding for CollegeGenerative AI CourseBecause an agent is only as good as the tool calls it makes, and those are easy to get wrong without seeing it.
The Liverpool project shows how a fluent, confident agent can return 5 or 59 when the answer is 49, without any error message. Nothing in the reply warns the user. The only defence is a person who knows to ask which field was searched, whether the match was exact, and whether the text was cleaned first.
Learners who have built the four versions themselves read agent output differently. They look for the query behind the answer, test edge cases like trailing spaces and similar names, and design tools that report how they reached a number. That is the difference between using AI agents and being misled by them.
In 2026 anyone can ask an agent a question; a Liverpool learner who can see inside the tool call is the one who knows whether to believe the answer. The longer argument is in why teenagers should still learn to code in 2026.
The council area covers about 306 square kilometres, and an online lesson reaches every part of it in exactly the same way.
No drive across the south-west at peak hour: a learner logs in from wherever they are and the whole lesson goes on coding.
Teachers plan around NSW stages and term dates, and lessons stop for the state school holidays.
The opening lesson is a proper class, finishing with a recommended course, level and weekly time. Nothing is paid on the day.
Groups of fifteen to twenty learners at a similar stage, or one-to-one teaching when that suits better.
Normally two lessons a week at a set time, around eight a month, with holidays agreed ahead.
A 6:30pm Liverpool lesson in summer begins at 1pm India time; when Sydney returns to standard time it begins at 2pm there.
Clock changes in Sydney
Daylight saving starts in Sydney on Sunday 4 October 2026 and finishes on Sunday 4 April 2027. India keeps one time all year, so every lesson time we give is in Sydney time and the India time shifts by an hour twice a year.
Three figures, all stated plainly.
A full lesson free, with a course, level and time recommended at the end.
A month of lessons, normally eight, alongside fifteen to twenty learners at the same level.
About eight lessons a month on the same rhythm, with one teacher and one learner.
Outside India we bill in US dollars, so there is one fee for every Liverpool household. Nothing is charged until the trial has fixed a course and a time; holidays, missed lessons and changing format are explained 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
From that a teacher can plan the first session. A Year 9 student might start by asking an AI to write a quiz game and then testing it; an adult might go straight to building the school-search tool.
The questions Liverpool families and adult learners tend to ask first.
QuickStats for the Liverpool local government area records 233,446 people at the 2021 Census, with a median age of 34, five years younger than New South Wales as a whole.
Vibe coding means describing what a program should do in everyday language and letting an AI write the first version. The skill we teach is the second half: reading that code, testing it and correcting it, so the result does what was intended.
An AI agent is a language model that can call tools, such as a search over a file or a calculator, decide what to do with the results, and keep going until it has an answer. Its answers are only as reliable as the tool calls it chooses.
The same question about government schools in Liverpool returns 5, 2, 7, 59 or 49 depending on whether the agent searches names, the suburb field, or the council field, and whether it cleans text and matches exactly. Only an exact match on the council field gives the full and correct 49.
Vibe coding starts from about eight with Scratch and AI helpers. Agent building is aimed at teenagers from 13 and adults, once they are comfortable writing and testing a little Python.
Families choose an after-school or weekend time at the trial, always given in Sydney time. India runs 5 hours 30 minutes behind Sydney in summer and 4 hours 30 minutes behind in winter.
Only tools whose own published rules allow the learner's age. Teenagers use tools that permit their age group, and services whose terms require adults are kept to adult classes.
No, and we do not claim one. Every lesson is live online and we have no premises in Australia. A learner needs a laptop or desktop with sound and a reliable connection; the phone number shown is Indian.
The first lesson is free. After that a group place is USD 100 a month for two live lessons a week, about eight a month, with fifteen to twenty learners, or USD 150 a month one to one on the same pattern. Course, format and time are settled before anything is paid.
Between five and ten learners, matched by level, pace and goal rather than age or postcode. If no group suits the time a learner needs, one-to-one lessons are offered.
Our Sydney-wide guide is coding classes in Sydney. For the thinking behind checking AI output, see learn to think, not just use AI tools, and for teenagers starting with AI, AI classes for teens in Australia. The state picture is on coding and AI classes in New South Wales.