AQA A-level Computer Science 7517 Help

In short

AQA 7517 splits three ways: an on-screen Paper 1 of two and a half hours at 40 per cent, a written Paper 2 of the same length at 40 per cent, and a practical project worth 75 marks and 20 per cent. Paper 1 is the unusual one. AQA issues preliminary material and a skeleton program in advance, in whichever of C#, Java, Python or VB.Net a school has chosen, and candidates then write, adapt and extend programs in an electronic answer document during the exam. Preparing for it means reading and navigating a codebase somebody else wrote, which is a distinct skill from writing one from scratch and the one most students have least practice in. We teach both, plus the theory Paper 2 rewards. The first lesson is free, then USD 100 a month in a group or USD 150 one to one.

Start here

Three courses for A-level candidates

Depth in a language, the algorithms Paper 2 leans on, and the data work the project usually needs. Cards open the syllabus.

AQA A / 01

Python, Zero to Advanced

Depth well past GCSE, including the file handling, structures and object work a skeleton program will assume.

Open the syllabus →

Data Structures & Algorithms Course course thumbnail

AQA A / 02

Data Structures and Algorithms

Trees, graphs, complexity and the theory of computation content, worked on paper before any code.

Open the syllabus →

AQA A / 03

Data Science, Complete

For students whose project involves real data, and for the Big Data content in the specification.

Open the syllabus →

The skeleton program

You meet the code before the exam, and you did not write it

This is the defining feature of AQA's A-level and the one students most often under-prepare for.

What the specification says

For Paper 1, "We will issue Preliminary Material, a Skeleton Program (available in each of the Programming Languages) and, where appropriate, test data, for use in the exam." Candidates "answer a series of short questions and write/adapt/extend programs in an Electronic Answer Document provided by us".

Why this is a different skill

Writing a program from an empty file and extending somebody else's program are not the same activity. The second demands that you find your way around unfamiliar structure, work out what a function is for from how it is used, and change one part without breaking three others.

A student who has only ever written their own code from scratch meets that for the first time in an exam, which is the worst possible place to meet it.

What good preparation looks like

Read the skeleton out loud, function by function, and say what each one is for before changing anything. Draw the call structure on paper. Run it and break it deliberately to see what depends on what.

Then practise small modifications against a clock. Adding a feature to somebody else's program in twenty minutes is a skill that improves quickly and only with repetition.

Three habits the skeleton rewards
Habit	What it looks like	Why the exam rewards it
Reading before typing	Explaining each subroutine in your own words, without changing anything	The questions assume you know where things are; hunting in the exam costs minutes you do not have
Tracing by hand	Following a value through the program on paper	Short questions frequently ask what the program does with a given input
Small safe changes	Extending behaviour without disturbing what already works	Adapt and extend are the specification's own words for what Paper 1 asks

The skeleton is provided in each supported language, so a candidate works in whichever one their school picked. AQA supports C#, Java, Python and VB.Net, notes that Python 2 is no longer supported, and says schools "will be asked to indicate their programming language preference at the start of the study of the" course. That decision is made for a student, usually before they arrive in Year 12.

Source, read 20 September 2026: AQA AS and A-level Computer Science (7516, 7517) specification, version 1.6, 31 July 2023, for exams June 2016 onwards. Check the current version on AQA's site; your school will confirm which applies and which language it has chosen.

How it is assessed

Forty, forty, twenty

Three components, two of them exams of identical length and very different character.

AQA 7517 at a glance
Component	Format	Weight	What it assesses
Paper 1	On-screen exam, 2 hours 30 minutes	40 per cent	The ability to program, plus theory from the programming, data structures, algorithms and theory of computation sections, and the systematic approach to problem solving
Paper 2	Written exam, 2 hours 30 minutes	40 per cent	Data representation, computer systems, organisation and architecture, consequences of computing, networking, databases, Big Data and functional programming
Project	Non-exam assessment, 75 marks	20 per cent	Using the knowledge and skills from the course "to solve or investigate a practical problem", following a systematic approach

Paper 1 is not a written paper

It is sat on screen, in an electronic answer document AQA provides. A student who has revised programming only on paper is preparing for the wrong medium.

Paper 2 is genuinely broad

It ranges from logic gates to databases to functional programming, and the extended-answer questions reward writing that is structured rather than merely correct.

The project is a fifth of it

Twenty per cent and 75 marks, and unlike the papers it is spread over months, which makes it the component most often left too late.

Two A-level topics catch GCSE-strong students out because they have no GCSE equivalent at all: functional programming and Big Data. Neither is hard once taught, and both are frequently squeezed into the last term because they sit late in the specification. Planning them earlier is one of the cheapest improvements available to a Year 12.

On the project itself, the marks and the board-specific rules are set out on the A-level coursework page, which compares AQA and OCR directly. We do not work on any candidate's project, and that line does not move.

Preparing for each part

Three components, three different plans

The most common A-level mistake is revising all three the same way. They reward almost nothing in common.

What each component actually needs
Component	The highest-value habit	The commonest mistake
Paper 1	Working inside somebody else's codebase, weekly, at a keyboard	Practising only from a blank file, so the skeleton is unfamiliar in May
Paper 2	Writing extended answers and having them marked by a person	Revising by reading notes, which shows knowledge but never tests expression
Project	Starting early and keeping a decision log from day one	Beginning in Year 13 and writing the analysis after the code

For Paper 1

Use a codebase that is not yours every single week. Any real program will do: read it, explain it, extend it, break it. The specific skeleton arrives when AQA issues it, and by then the skill should already exist.

Then work at the keyboard, in the language your school chose, with the clock running. Fluency and navigation are the two things that decide this paper.

For Paper 2

The content is broad and mostly learnable, so the marginal marks are in extended answers. Write them, hand them to somebody who will mark them honestly, and rewrite.

Functional programming and Big Data deserve a term of their own rather than a fortnight in April.

Our line on the project is the same as everywhere on this site. We do not work on it, we do not review it, and we will decline if asked, because a project that counts for 20 per cent has to be the candidate's own. What we do teach is everything it assumes: structuring a program, testing it properly, working with data, and writing about your own decisions clearly. The page on understanding the code covers what the awarding bodies now say about AI in assessed work, which matters more at A level than anywhere else.

For students weighing up A level in the first place, or arriving from a different GCSE board, the three English GCSE specifications differ more than most people expect, and the AQA GCSE page is the natural place to start.

Progression

Four rungs across Years 12 and 13

Where a student stands is judged by what they can do inside an unfamiliar program, not by what they can recite.

What an A-level candidate should be able to do
Rung	When	What should be true
1. Fluent beyond GCSE	Early Year 12	Comfortable with structures, files and objects in the school's chosen language
2. Reading other people's code	Across Year 12	Able to explain an unfamiliar program and extend it without breaking it
3. The hard theory	Year 12 into Year 13	Theory of computation, complexity, functional programming and Big Data actually taught, not deferred
4. Writing to be marked	Year 13	Extended answers that are structured, and a project with a decision log behind it

Rung two is the differentiator

It is the rung nobody sets out to climb, because school work is almost always writing your own program from nothing. The exam is not.

The topic sequence past A level runs down the coding roadmap.

Rung three cannot be left to spring

Functional programming and Big Data have no GCSE ancestor, so a student meets them cold. Late exposure turns learnable content into a panic.

Lessons ease off before the papers and restart afterwards.

The catalogue

Nine courses for sixth-form computer scientists

Grouped by which part of the A-level they serve. The free lesson settles the starting point.

I

Depth in a language

For the on-screen paper

UK / A1 / 01

Python, zero to advanced

Well past GCSE, into the structures and object work a skeleton program assumes.

Open the syllabus

UK / A1 / 02

Java, complete

For sixth forms that chose Java, taught with the same emphasis on reading as writing.

Open the syllabus

UK / A1 / 03

Git and GitHub

Navigating and changing a codebase safely, which is the exact Paper 1 skill in professional form.

Open the syllabus

II

The theory

For the written paper

UK / A2 / 01

Data structures and algorithms

Trees, graphs and complexity, worked on paper where the exam will ask for them.

Open the syllabus

UK / A2 / 02

Databases and SQL

The databases section done properly rather than skimmed in the final term.

Open the syllabus

UK / A2 / 03

Competitive programming

Speed and correctness under constraint, for students who want to be well clear of the syllabus.

Open the syllabus

III

Around the project

Skills it assumes, never the project itself

UK / A3 / 01

Data science, complete

For projects built on real data, and for the Big Data content in the specification.

Open the syllabus

UK / A3 / 02

Full stack web development

Building something with a database behind it, which many projects turn out to need.

Open the syllabus

UK / A3 / 03

Python and AI automation

Working alongside tools properly, with a record of what came from where.

Open the syllabus

How lessons work

Somebody else's code, every week

Teachers work from India, where the clock never moves, so the UK is five and a half hours behind from October and four and a half from March. Slots are agreed in UK time.

Early evening

The usual choice for Year 12 and 13 students.

Later evening

For students with sport, work or a long journey home.

Weekend morning

Long enough for full timed practice and a proper review.

Read before you write

Every week a student explains a program they did not write, function by function, before changing a line of it.

At a keyboard

Paper 1 is on screen, so practice is on screen, in the language the school chose.

Extended answers marked

Written responses handed in and marked by a person, because that is the only way expression improves.

Theory taught early

Functional programming and Big Data planned into Year 12 rather than left to the final term.

A clear line on the project

We never work on it, review it or advise on its content. We teach the skills it assumes.

Groups by level

Five to ten students at one stage, reading each other's changes to the same program.

Student work

Work our older students have published

Four projects built in lessons, none of them anybody's assessed coursework. More in student labs.

AI and ML

NutriLife

An AI nutrition coach that reads what you eat and works you toward a target.

by Bhavya · Open it

AI and ML

Misti

A chatbot that answers mathematics and programming questions, built and deployed by a student.

by Harshit · Open it

AI and ML

GuardianX

An assistant that helps a young person recognise unsafe situations online.

by Vivaan · Open it

Web app

SkyCast

A weather forecasting site with live conditions for any location.

by Krish

Fees

What A-level tuition costs

One charge a month in US dollars, the rate for every country outside India. The free lesson comes first and billing starts only when a course and a weekly time are agreed.

Free first class

USD 0

no card required

A full lesson on real 7517 material
An honest view of which component is weakest
Nothing asked for but a number

Book it

Group batch

USD 100

a month, billed in US dollars

Five to ten students at one stage
The same teacher across the year
Written answers marked weekly
Fewer lessons in the run-up to the papers

Start here

One to one

USD 150

a month, billed in US dollars

Only the student and the teacher
Planned backwards from the exam timetable
Suits a candidate with a specific grade in view

Enquire

What families say

Rated 4.9 across 547 Google reviews

From Google, word for word, unpaid and unprompted.

★★★★★

"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

Questions about AQA 7517

What A-level candidates and parents ask

How is AQA A-level Computer Science assessed?

Paper 1 is an on-screen exam of two and a half hours worth 40 per cent, Paper 2 is a written exam of the same length worth 40 per cent, and the non-exam assessment is 75 marks and 20 per cent.

What is the skeleton program?

A program AQA issues before Paper 1, available in each supported language, alongside preliminary material and, where appropriate, test data. Candidates work with it during the exam.

Is Paper 1 taken on a computer?

Yes. Students answer short questions and write, adapt and extend programs in an electronic answer document AQA provides.

Which languages does AQA support?

C#, Java, Python and VB.Net, with Python 2 no longer supported. Schools indicate their preference at the start of the course, so the choice is usually made for a student.

How should a student prepare for the skeleton?

By working inside other people's programs every week, long before the skeleton arrives. Reading, explaining, extending and breaking code is a separate skill from writing your own.

What is in Paper 2?

Data representation, computer systems, organisation and architecture, consequences of computing, networking, databases, Big Data and functional programming, in compulsory short-answer and extended-answer questions.

Which topics catch students out?

Functional programming and Big Data, because neither has a GCSE equivalent and both sit late in the specification, so they are often rushed in the final term.

Will you help with the project?

No. It is 20 per cent of the A-level and it has to be the candidate's own. We teach the skills it assumes and decline any request to work on the project itself.

Can a student use AI while preparing?

For our own exercises, yes, in a fixed order: understand, predict, then ask, with a record of what the tool contributed. For assessed work the awarding bodies' rules apply and they are strict.

What does it cost?

The opening lesson is free. After that a group place is USD 100 a month and one-to-one teaching USD 150, billed monthly with nothing in advance.

Elsewhere on this site

Where to read next

Each page takes a different part of this further.

A-level Computer Science coursework

Where the project marks sit at AQA and OCR, and what the boards say about AI.

AQA GCSE Computer Science 8525

The same board two years earlier, where the difficulty is notational.

Understand the code, do not copy-paste it

Why declared AI content cannot earn marks in assessed work.

Coding and AI classes in England

What the computing curriculum requires and who has to follow it.

British Informatics Olympiad

Where a strong A-level student can go beyond the specification.

Coding classes in the UK

The national page, and the index of every UK page we publish.

Start here

Book a free first lesson

Send a number and we will reply at a British hour you choose. The lesson uses real A-level material, at a keyboard, and finishes by naming the weakest of the three components.

Prefer reading? Each course page publishes its syllabus, how we teach covers the method and its limits, and the coding roadmap sets out the sequence.

WhatsApp us · +91 91233 66161 · contact@modernagecoders.com

WhatsApp is free from a British mobile and gets the fastest reply. The number is registered in India, which we say here rather than leave you to find.

Three courses for A-level candidates

Python, Zero to Advanced

Data Structures and Algorithms

Data Science, Complete

You meet the code before the exam, and you did not write it

Why this is a different skill

What good preparation looks like

Forty, forty, twenty

Paper 1 is not a written paper

Paper 2 is genuinely broad

The project is a fifth of it

Three components, three different plans

For Paper 1

For Paper 2

Four rungs across Years 12 and 13

Rung two is the differentiator

Rung three cannot be left to spring

Nine courses for sixth-form computer scientists

Depth in a language

Python, zero to advanced

Java, complete

Git and GitHub

The theory

Data structures and algorithms

Databases and SQL

Competitive programming

Around the project

Data science, complete

Full stack web development

Python and AI automation

Somebody else's code, every week

Read before you write

At a keyboard

Extended answers marked

Theory taught early

A clear line on the project

Groups by level

Work our older students have published

NutriLife

Misti

GuardianX

SkyCast

What A-level tuition costs

Rated 4.9 across 547 Google reviews

What A-level candidates and parents ask

How is AQA A-level Computer Science assessed?

What is the skeleton program?

Is Paper 1 taken on a computer?

Which languages does AQA support?

How should a student prepare for the skeleton?

What is in Paper 2?

Which topics catch students out?

Will you help with the project?

Can a student use AI while preparing?

What does it cost?

Where to read next

A-level Computer Science coursework

AQA GCSE Computer Science 8525

Understand the code, do not copy-paste it

Coding and AI classes in England

British Informatics Olympiad

Coding classes in the UK

Book a free first lesson

Keep exploring Modern Age Coders

Related courses and guides

Learn more

Free resources

From the blog

Start here