Which are the best AI and programming classes in Harrogate?
The ONS counted 75,515 people in the Harrogate built-up area at the 2021 census and 15,785 in Knaresborough, with Starbeck, Bilton, Jennyfield and Harlow Hill among the suburbs recorded around the town. From six-year-olds to adults of 67, anyone in the district can learn AI, programming, Python, vibe coding and maths on a live video link, taught by our India-based team one-to-one or in a group of five to ten sharing a level. We teach how to think before any tool, so learners stay in charge of the AI they use. The free first lesson ends with a course recommendation. Harrogate's project tackles a question people ask phones and AI assistants every day, "is it open now?", by parsing the real opening hours recorded for the town centre. Staying on after that is USD 100 per month in a shared class, or USD 150 per month with a tutor to yourself.
Ask a phone or an AI assistant whether a café is open and, behind the scenes, a program has to read a line like "Mo-Fr 08:00-17:00; Sa,Su 09:00-17:00" and work out the answer. OpenStreetMap stores opening hours for places in central Harrogate in exactly this format, and the format has an official specification. This project uses regular expressions in Python, the standard tool for matching text patterns, to teach a program to read those lines. The first attempt looks fine on tidy examples and fails on four out of five real ones, which is precisely the lesson: real data is always messier than the examples, and software, whether written by a person or an AI, has to be tested on the real thing.
Facts last verified 29 September 2026. Teaching is online; no Harrogate branch is claimed.
Four age-matched starting points, each opening with a no-cost live lesson that needs no card to reserve.

Our how-to-think programme: spotting patterns in text, rules and the exceptions to them.
See the syllabus
Scratch games, then little apps made by describing them to AI and testing each feature.
See the syllabus
Python and web projects with AI help, including the "is it open now?" parser.
See the syllabus
Python in depth: text processing, regular expressions, data 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 syllabusPublished ONS census counts and the neighbourhoods recorded around the town.
| Built-up area | People (2021) |
|---|---|
| Harrogate | 75,515 |
| Knaresborough | 15,785 |
| Pannal | 2,530 |
We show these ONS figures as published and do not add them together. Starbeck, Bilton, Jennyfield, Harlow Hill, Oatlands and New Park are recorded as suburban areas, and Killinghall as a village, in North Yorkshire. Schools here follow the national curriculum for England; tell us the holiday weeks and we will leave them free of lessons.
More options are on coding classes in North Yorkshire and Yorkshire and the Humber. Why every course starts with reasoning rather than prompting is on learn to think, not just use AI tools.
Write a pattern, test it on the town centre's real data, then read the specification and find what you missed.
From an OpenStreetMap API download covering the middle of Harrogate, the learner extracts every shop, café, restaurant and other amenity that has an opening_hours tag: 60 places, written in 55 different ways. The first regular expression matches the obvious pattern, a day range and one time range such as "Mo-Sa 10:00-17:30". It reads just 12 of the 60. Most places are more complicated: different hours on different days, lunchtime breaks, Sundays shown separately.
The second version splits the text at semicolons into separate rules, understands day lists like "Sa,Su" and ranges like "Mo-Fr", allows more than one time range in a day, and treats "off" as closed. It also follows a rule from the official specification on the OpenStreetMap Wiki: when two rules cover the same day, the later one wins, or in the specification's words, "the earlier rule is overridden". This parser reads 54 of the 60 places.
| Approach | Places read |
|---|---|
| One simple regular expression | 12 of 60 |
| Rule-by-rule parser for the common cases | 54 of 60 |
| Left over: public holidays marked PH | 2 |
| Left over: extra rules joined by a comma | 2 |
| Left over: the word "closed" | 1 |
| Left over: a time with no days given | 1 |
The six left over are not mistakes in the data. Checking the specification shows that public holidays, the word "closed" and extra rules joined by a comma are all part of the official format; the parser simply does not handle them yet. That is a lesson every programmer meets: a format is always bigger than the examples you first looked at, and the specification is the place to find out how much bigger.
Getting it wrong has a real cost. Of the 54 places the parser can read, 42 list more than one rule. A lazy program that reads only the first rule gives the wrong open-or-closed answer for 755 of the 9,072 hours in a week across those places, 8.3%. With the full parser, 17 of the 54 are open at half past four on a Sunday afternoon and 10 at eight on a Monday evening.
Read opening-hours signs, write them in a shared code, and spot which ones break the code.
Write a first regular expression in Python and count how many real entries it matches.
Build the rule parser, test "open now" across the week and extend it from the specification.
Opening hours are from OpenStreetMap and its contributors under the Open Database Licence, and the format is defined on the OpenStreetMap Wiki. The parser, the counts and the open-now checks are our own work; we do not name individual businesses.
Passing on tidy examples proves little; passing on real data proves more.
| In the opening-hours project | When AI writes or runs code |
|---|---|
| One regex read 12 of 60 | A neat-looking answer can fail on most real inputs |
| The parser still missed 6 | Keep a list of cases the code cannot handle |
| The specification settled it | Check the official source, not just examples |
| First-rule-only was wrong 8.3% of the time | Small shortcuts create real wrong answers |
| Later rules override earlier ones | Order and precedence matter in rules |
Ask an AI assistant for a regular expression that reads opening hours and it will usually give you something like the first attempt: tidy, confident and wrong for most of the town. In our vibe coding lessons, where the learner describes a program and the AI drafts it, Harrogate learners run every draft against the real 60 entries before trusting it. AI agents that answer questions such as "is it open now?" for you depend on exactly this kind of parsing, so a wrong rule turns into a wasted journey. Learners reach agents once Python feels easy, typically late in their teens or as adults, and anything involving Copilot Studio is taught privately. You can read about that route on the page for UK students learning to build agents, and our thinking on it in understand the code, don't copy-paste.
OpenStreetMap, the ONS and the businesses in the data are all independent of Modern Age Coders; we only used openly published records, and the parser, flaws included, is our responsibility.
We start from the school year, then the free lesson places the learner properly.
Patterns, rules and the exceptions that break them.
Problem Solving and Computational Thinking for KidsScratch Coding for KidsGames and small apps built with AI help and checked by the learner.
Vibe Coding for KidsPython and AI for KidsStrings, regular expressions and testing alongside GCSE and A level.
Vibe Coding for TeensAI and Machine Learning for TeensText processing, data and AI agents, built properly in Python.
Python MasterclassGenerative AI CourseIt is a compact pattern for matching text, and AI can write one, but only testing shows whether it works.
In Harrogate the obvious regular expression read 12 of 60 real entries. An AI would likely have produced something similar, and it would have looked just as convincing.
Learners who have tested patterns on real data, and then checked the specification, know to ask for evidence before trusting any code, including code an AI wrote.
A Harrogate teenager who can test code against messy real data will use AI tools with real confidence, and that is a strong reason to learn to code in 2026. The longer argument is in why teenagers should still learn to code in 2026.
Any home with a computer and broadband good enough for video can join.
Students write, prompt and run their programs themselves, while the tutor follows on screen share and asks questions.
The free lesson shows where to begin, whatever the school year, and any exam board is noted.
No fee for lesson one, which closes with our suggestion of a course.
A class brings together between five and ten learners from around Britain who have reached the same point.
Lessons stop for the school holidays.
Our tutors follow the UK clock changes, so your lesson hour does not move.
Why we teach online
Five learners at the same stage and free at the same hour are unlikely to live near each other. Online, that does not matter.
Harrogate learners pay our international rate, which applies in every country except India.
A complete first lesson free of charge, with a course suggestion at the end.
Roughly eight live small-group lessons a month.
Roughly eight live one-to-one lessons a month.
Fees are charged in US dollars rather than pounds. Invoicing begins after the free lesson, once a course and a regular time are agreed; see the pricing page for holidays, missed sessions 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 a year group or age and what the learner is into. Possible trial activities: a pattern-hunting puzzle, an AI-assisted Scratch game, a first go at Python, or a mini parser for real text.
The opening-hours project, regular expressions, vibe coding and practical points.
The ONS gives 75,515 for the Harrogate built-up area at the 2021 census, and 15,785 for Knaresborough.
Yes, live online, for learners aged 6 to 67 in Harrogate, Knaresborough and the villages around them.
A short pattern that describes text to find or check, such as "a day range followed by a time range"; most programming languages, including Python, support them.
Learners parse the real opening hours of places in central Harrogate from OpenStreetMap, test their code against the official specification and answer "is it open now?" for every hour of the week.
Yes, at any age: the learner decides what to build, an AI drafts it, and the learner tests every part.
As soon as they are comfortable in Python, which for most means the later teens or adulthood; Copilot Studio agent lessons are private only.
No, all lessons are live online.
Computer science and maths are both covered, aimed at genuine understanding rather than a promised grade.
Lesson one: nothing. Afterwards: USD 100 monthly in a class of peers, or USD 150 monthly on your own.
No, they pause. Send us the dates.