Frequently Asked Questions

PaperGym is an interactive learning platform designed to help you master modern AI by implementing research papers through hands-on coding drills.

Instead of just reading papers, you'll:

• Complete bite-sized coding challenges that implement key concepts
• Get instant feedback from automated tests
• Access progressive hints when you're stuck (3 levels per drill)
• Track your progress as you master each paper

Learning on PaperGym is simple and effective:

1. Browse papers - Choose from 50+ curated AI research papers organized by topic
2. Read the summary - We provide concise explanations and context for each paper
3. Complete coding drills - Implement key concepts in bite-sized coding exercises
4. Get instant feedback - Automated tests tell you immediately if your solution is correct
5. Use hints if needed - Access up to 3 progressive hints per drill (Free: 3 hints/week, Premium: unlimited)
6. Earn points and track progress - Build momentum as you complete more drills

We cover a wide range of modern AI topics, including:

• Transformers & Attention - "Attention Is All You Need", BERT, GPT, and more
• Computer Vision - CNNs, ResNets, Vision Transformers, object detection
• Natural Language Processing - Word embeddings, sequence models, language models
• Reinforcement Learning - Q-learning, policy gradients, DQN
• Generative Models - VAEs, GANs, diffusion models
• Optimization & Training - Batch normalization, dropout, advanced optimizers

We add new papers regularly based on community feedback and the latest research trends.

PaperGym is perfect for:

• ML Practitioners - Deepen your understanding of foundational papers and stay current with the field
• Students - Learn research concepts through implementation rather than just theory
• Software Engineers - Prepare for machine learning interviews with hands-on practice
• Researchers - Explore new domains by implementing papers outside your specialty
• Self-learners - Build practical ML skills at your own pace

You should have basic Python programming experience and familiarity with fundamental ML concepts (linear algebra, calculus basics).

No, you don't need to read the full paper to complete the drills. We provide:

• Concise summaries that explain key concepts
• Context for why the paper is important
• Focused drills that implement the core ideas
• Links to the original papers for those who want to dive deeper

That said, reading the original papers can provide valuable context and deeper insights. We encourage you to explore the papers that interest you most.

Free Tier includes:

• Access to ALL 50+ research papers
• Complete ALL coding drills and learning paths
• 3 LLM-powered hints per drill (for all drills)
• Basic progress tracking
• Auto-save for 7 days

Premium Tier ($4.99/month or $35.88/year) adds:

• Unlimited LLM hints - Never get stuck, learn faster
• Advanced analytics - Detailed insights into your learning progress
• Unlimited auto-save - Your work is always saved
• Priority support - Faster response times

Premium is incredibly affordable:

• Monthly Plan: $4.99/month - billed monthly
• Annual Plan: $35.88/year (equivalent to $2.99/month) - Save 40%

That's less than the cost of a coffee per month for unlimited learning support. Plus, we offer a 7-day money-back guarantee on all new Premium subscriptions.

View detailed pricing →

You can cancel your Premium subscription at any time with no penalties:

1. Go to your Account Settings
2. Click on "Manage Subscription"
3. Select "Cancel Subscription"

When you cancel:

• You'll retain Premium access until the end of your current billing period
• No further charges will be made
• You can reactivate your subscription anytime

Need help? Email us at [email protected]

We offer a 7-day money-back guarantee on all new Premium subscriptions.

If you're not satisfied with Premium for any reason, simply:

1. Email us at [email protected] within 7 days of your purchase
2. We'll process a full refund, no questions asked

Important notes:

• The 7-day guarantee applies only to the first payment of a new subscription
• Renewal payments are not eligible for refunds
• Refunds are processed within 5-7 business days

We don't currently offer a separate student discount, but we're considering it for the future.

However, our Free tier is generous and perfect for students:

• Access to ALL papers and drills
• 3 LLM hints per drill (enough for most learners)
• No credit card required

If you need Premium features and cost is a concern, please email us at [email protected]. We evaluate requests on a case-by-case basis.

Currently, all drills are in Python, using popular ML libraries:

• NumPy - For numerical computing and array operations
• PyTorch - For deep learning implementations
• Standard Python libraries (math, collections, etc.)

We chose Python because it's the lingua franca of machine learning and most research papers provide reference implementations in Python.

Support for additional languages (JavaScript, Julia, etc.) is on our roadmap based on community demand.

No installation required!

PaperGym runs entirely in your browser:

• In-browser code editor with syntax highlighting and autocomplete
• Cloud-based code execution in a sandboxed environment
• Works on any device with a modern web browser (Chrome, Firefox, Safari, Edge)
• No need to install Python, PyTorch, or any dependencies

Just sign up and start learning immediately. All you need is an internet connection.

Your code runs in a secure, sandboxed cloud environment:

• Isolated execution - Each submission runs in its own container
• Resource limits - CPU, memory, and time limits prevent abuse
• No network access - Code cannot access external resources
• Automated cleanup - Environments are destroyed after execution

This ensures:

• Your code is executed safely without affecting other users
• Consistent, reproducible results
• Fast feedback (most tests run in under 10 seconds)

We provide multiple levels of support to help you learn:

1. Progressive Hints (3 levels):

• Level 1: Gentle nudge in the right direction
• Level 2: More specific guidance with pseudocode or key insights
• Level 3: Detailed explanation with implementation approach

2. Test Feedback:

• See which test cases pass or fail
• Review error messages and assertion failures

3. Paper Context:

• Re-read the paper summary
• Check the original paper for additional context

Hint limits: Free tier gets 3 hints per drill, Premium gets unlimited hints.

Yes! Once you successfully complete a drill, you can view our reference solution.

This allows you to:

• Compare your approach with ours
• Learn alternative implementations
• Understand best practices and optimizations
• See how to write cleaner, more efficient code

Solutions are only available after you've completed the drill to encourage learning through implementation rather than copying.

Your progress is tracked automatically across multiple dimensions:

Dashboard Overview:

• Total points earned
• Number of drills completed
• Papers mastered
• Current learning streak

Per-Paper Progress:

• Completion percentage for each paper
• Which drills you've completed
• Drills remaining

Premium Analytics (Premium only):

• Detailed insights into your learning patterns
• Time spent on different topics
• Difficulty analysis
• Personalized recommendations

Retrying drills: Yes! You can retry any drill at any time to practice or improve your solution. However, points are only awarded once per drill.

Resetting all progress: This feature is not currently available. If you have a specific need to reset your account, please contact support at [email protected].

We recommend keeping your progress intact as it shows your learning journey and helps us provide better recommendations.

Yes! Your code is automatically saved as you work:

• Auto-save frequency: Every few seconds while you type
• Free tier: Code saved for 7 days
• Premium tier: Unlimited auto-save retention

This means you can:

• Close your browser and resume where you left off
• Switch between devices
• Never lose your work due to browser crashes

Your submitted solutions (passed drills) are saved permanently for both Free and Premium users.

Absolutely! PaperGym is excellent for ML interview preparation:

How it helps:

• Hands-on implementation - Practice coding ML algorithms from scratch
• Foundational papers - Understand the "why" behind common ML techniques
• Immediate feedback - Know right away if your implementation is correct
• Progressive difficulty - Build from basics to advanced topics

What interviewers ask about:

• Implementing attention mechanisms, transformers
• Building neural networks from scratch
• Understanding optimization algorithms
• Explaining tradeoffs in model architectures

Many of our users have successfully used PaperGym to prepare for ML roles at top tech companies.

We're here to help! You can reach our support team at:

When contacting support, please include:

• Your account email
• A description of the issue
• Any relevant screenshots or error messages

Premium subscribers receive priority support with faster response times.