Which are the best AI and programming classes in Ormeau, Belfast?
The Belfast ward named Ormeau had 6,532 usual residents at the 2021 census, and the neighbouring-named Rosetta ward 5,793, according to NISRA; OpenStreetMap labels Lower Ormeau, Ballynafeigh and Ravenhill as suburbs in the same part of the city. Tutors working from India teach AI, programming, Python, vibe coding and maths over live video to anyone from six to 67, privately or in a class of five to ten pitched at one level. We teach how a method works before how to call it, so learners can tell what an AI grouping really means. Lesson one is free and ends with our course advice. The Ormeau project clusters 341 shops and cafes two ways and asks whether the centre of a group should be an average point that may sit in the middle of a road, or a real place you could walk to. After the trial, a class is USD 100 a month and private tuition USD 150 a month.
Clustering is how AI sorts things into groups without being told the answer, and the usual method, k-means, describes each group by its average. Averages are convenient, but the average of forty shop locations is not a shop; it may be a back garden. A close relative called k-medoids makes one change: the centre of each group must be one of its actual members, the one with the smallest total distance to the rest. That member is called the medoid, or exemplar. The change brings two bonuses: each group has a real representative you can name, and the method works with any measure of distance, including how far it is on foot. Ormeau's shops and cafes make a good test.
Facts last verified 30 September 2026. Teaching is online; no Ormeau branch is claimed.
Choose the row that fits the learner's age. A free live lesson opens every course, and nothing is paid to book.

The how-to-think course: sorting into groups and choosing one member to stand for each.
See the syllabus
Scratch games designed by the learner, made with an AI and tested carefully.
See the syllabus
Machine learning in Python, with k-means and k-medoids on real Ormeau map data.
See the syllabus
Python for data science, clustering, machine learning and AI agents.
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 syllabusNISRA census figures for areas that carry these names, and how the map labels the district.
| Area | Residents (2021) |
|---|---|
| Ormeau ward | 6,532 |
| Rosetta ward | 5,793 |
| Botanic district electoral area | 49,727 |
These are three different official areas and their figures are not added or compared as parts of one another; "Ormeau" in everyday use does not match the ward line. Within our study rectangle OpenStreetMap labels Lower Ormeau, Ballynafeigh, Ravenhill, Botanic and Stranmillis. Schools teach the Northern Ireland Curriculum from P1 to Year 14. Give us the school holiday dates and lessons will be arranged round them.
See Belfast, Malone and CCEA A level Software Systems Development help. Our reasons for teaching reasoning first are on learn to think, not just use AI tools.
The same 341 places, grouped two ways, then grouped again by walking distance.
The learner downloads OpenStreetMap data for a rectangle covering the Ormeau Road and the streets around it and keeps the shops, cafes, restaurants, takeaways, pubs and bars mapped as points: 341 of them. Both methods are asked for five groups. k-means moves its five centres to the average position of each group. k-medoids instead tries each member as the centre and keeps the one with the smallest total distance to the others. Each method is restarted 20 times and its tightest result kept.
| Measure | k-means | k-medoids |
|---|---|---|
| Is each centre a real place? | No: 6.5 to 97.3 m from the nearest one | Yes, all five |
| Average distance from a place to its centre | 303.7 m | 291.8 m |
| Typical shift of a centre when the data is resampled | 48.0 m | 36.0 m |
| Shift in the worst tenth of resamples | 156.6 m | 104.5 m |
None of the k-means centres lands on a real place; one is almost 100 m from the nearest shop. Every medoid is a place with a name and a door. The medoids also move less when the data is disturbed: redrawing the 341 places at random 60 times shifts a typical k-means centre by 48.0 m and a typical medoid by 36.0 m. And because k-medoids only needs a table of distances, the learner can swap in walking distance. Here the median walk between two places is 1,598 m against 1,305 m in a straight line. With six groups, choosing medoids by walking distance cuts the share of places more than a kilometre on foot from their centre from 2.1% to 0.9%, though the two groupings agree on 96.4% of pairs, so in this compact district the straight line was already a fair guide.
Sort picture cards into groups, then pick the one card that sums up each group and defend the choice.
Run k-means on Ormeau's shops in Python and check what is actually at each centre.
Code k-medoids, test its stability by resampling, and switch from straight-line to walking distance.
Shops, cafes and paths are from OpenStreetMap and its contributors under the Open Database Licence. The groupings, distances and stability tests are our own work. No business is named, and a group on this page is a statistical convenience, not a neighbourhood boundary.
A group is easier to trust when you can point at the thing that stands for it.
| In the k-medoids project | When AI groups data for you |
|---|---|
| k-means centres were not real places | An average may describe nothing that exists |
| Medoids were actual shops or cafes | Ask for a real example of each group |
| Medoids moved less under resampling | Check whether the answer survives small changes |
| Walking distance changed a few groups | The distance measure is a modelling choice |
| 96.4% of pairs stayed together | Report how much a choice mattered, not only that it did |
AI tools cluster customers, documents and images all the time, and they usually describe each cluster with an average nobody can inspect. Asking for an exemplar, a real member that typifies the group, makes the result something a person can check. In vibe coding the learner tells an AI what to build and the AI writes it; our Ormeau learners follow up by asking which distance it used and what sits at each centre. Agents that sort and summarise data for you should be asked the same. We leave agent projects until Python is no longer the hard part, typically for sixth-formers and adults, and Copilot Studio work is always taught privately. See AI agents: the course for UK students and understand the code, don't copy-paste.
We have no connection with OpenStreetMap or NISRA beyond using their open data; the clustering work, including any errors, is our own.
The school year, P1 to Year 14, suggests a starting point and the free lesson confirms it.
Grouping, typical examples and explaining a choice.
Problem Solving and Computational Thinking for KidsScratch Coding for KidsGames and apps planned by the learner, built with an AI, then tested.
Vibe Coding for KidsPython and AI for KidsClustering, distance measures and stability alongside CCEA GCSE and A level.
AI and Machine Learning for TeensStatistics & ProbabilityUnsupervised learning, evaluation and AI agents in Python.
Python MasterclassGenerative AI Coursek-medoids groups data around real members, called medoids, chosen to have the smallest total distance to the rest of their group, whereas k-means groups around averages that need not be real points; k-medoids also works with any distance measure.
On 341 shops and cafes around the Ormeau Road, no k-means centre was a real place (one was 97.3 m from the nearest), every k-medoids centre was, and under resampling the medoids shifted a typical 36.0 m against 48.0 m.
Learners who have compared the two ask of any AI grouping: what real example stands for each group, and which distance decided it?
Being able to ask for the exemplar gives Ormeau teenagers a practical way to check AI groupings, and they learn it by coding both methods themselves. The longer argument is in why teenagers should still learn to code in 2026.
Needed at home: a computer, a webcam and broadband that can manage a video call.
Code and prompts are the student's own work, with the tutor on screen share asking what each output means.
The free lesson shows us the level; CCEA courses are noted where they apply.
Lesson one is free and finishes with a course recommendation.
Each class is five to ten learners at a shared level, from anywhere in the UK.
Term time only.
Tutors shift with the UK clocks, so your slot never moves.
Why online
A class needs five people at one level free at one time, which a single district rarely supplies. Online, the class is drawn from a whole country.
Ormeau learners are on our international rate, the same for every country apart from India.
A free full-length lesson and then our advice.
About eight live group lessons every month.
About eight live private lessons every month.
We charge in US dollars and publish no sterling prices. The first bill comes only after the trial has fixed a course and a weekly time, and holidays, absences 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
Let us know an age or Year group and one thing the learner enjoys. We might begin with a card-sorting puzzle, a Scratch game built with an AI, a short Python script, or grouping real places on a map.
Clustering, medoids, the shop project, vibe coding and practical details.
NISRA counted 6,532 usual residents in Ormeau ward at the 2021 census. The everyday area called Ormeau is not an official unit and has no figure of its own.
They are. Each lesson is a live video call, so Ballynafeigh, Ravenhill and every other part of Belfast can join, from age 6 to 67.
The member of a group with the smallest total distance to all the other members. Unlike an average, it is always a real item from the data.
When you need each group represented by a real example, when outliers would drag an average, or when your distance is not a straight line, such as walking distance or a similarity score.
Clustering 341 mapped shops and cafes with k-means and k-medoids, testing how stable each is under resampling, and repeating k-medoids with walking distance.
Yes, for every age: learners describe the program, an AI writes a draft, and they test and repair it.
When Python has stopped being the obstacle, which tends to be sixth form or later. Copilot Studio is taught privately.
There is, for GCSE and A level Digital Technology, Software Systems Development and Maths, taught for understanding, with no grade promised.
The trial costs nothing. After it, a class place is USD 100 each month and a private tutor USD 150 each month.
Yes. Tell us the dates and we skip them.