- Implement scaled dot-product attention with masking support
- Build multi-head attention with learnable projections
- Create sinusoidal positional encoding for sequence understanding
- Add layer normalization for training stability
- Complete transformer block with residual connections
- Include self-attention wrapper and utility functions
- Full inline testing with 100% pass rate
- Educational content explaining attention mechanisms
- Foundation for modern AI architectures (GPT, BERT, etc.)
This module bridges classical ML (tensors, layers, networks) with
modern transformer architectures that power ChatGPT and contemporary AI.
- Emphasize always working in dev branch
- Main branch for stable releases only
- Recommend feature branches for all changes
- Add YAML-style description for Cursor
- Clear workflow steps and quick reference
- Address math anxiety: explain math learning approach
- Address validation fears: highlight testing and feedback
- Address flexibility concerns: explain module dependencies
- Address toy project skepticism: emphasize real data and results
- Focus on actual questions students ask vs generic course info
- 4 key questions for students already interested in the course
- Focus on practical learning concerns vs skepticism
- Shorter than GitHub FAQ - appropriate for committed learners
- Covers time investment, skill level, support, modern relevance
- Remove career projections and salary mentions (too sales-y)
- Add dropdown format for compact presentation
- Logical order: basic skepticism → advanced concerns → practical details
- Focus on learning benefits and technical substance
- More concise and scannable format
- Address Transformer dominance vs foundations learning
- Explain why not just use PyTorch/TensorFlow
- Differentiate from basic tutorials - emphasize systems thinking
- Show concrete ROI and career impact
- Bridge academic vs practical concerns
- Provide realistic time investment and career paths
- Address common objections with evidence-based responses
- Replace dry text description with engaging Mermaid flowchart
- Show clear progression through 4 educational layers: Foundation → Deep Learning → Production → Mastery
- Use color coding and visual flow arrows to demonstrate module dependencies
- Make it immediately clear how each module builds into the next
- Regenerated all chapters with YAML-based difficulty ratings
- Updated book with improved navigation and fixed appendix links
- Applied copyright year 2025 across all pages
- Integrated inclusive language changes throughout generated content
- Book now reflects all UX and consistency improvements
- Added educational metadata (difficulty, time_estimate) to all module.yaml files
- Updated convert_readmes.py to read from YAML instead of hardcoded mappings
- Standardized difficulty progression: ⭐ → ⭐⭐ → ⭐⭐⭐ → ⭐⭐⭐⭐ → ⭐⭐⭐⭐⭐🥷
- Fixed path resolution for YAML reading in book build process
- Eliminated duplication: single source of truth for educational metadata
- Capstone gets special ninja treatment (⭐⭐⭐⭐⭐🥷) as beyond-expert level
- Added detailed file hierarchy showing modules/source/, tinytorch/, book/, tito/ organization
- Included workflow explanation from development to testing to deployment
- Added difficulty progression visualization (⭐ to ⭐⭐⭐⭐⭐🥷)
- Enhanced module descriptions with clear learning objectives
- Improved onboarding experience for new contributors and students
- Updated book generation to include 15_capstone with 5-star difficulty rating
- Changed time estimate from '20-40 hours' to 'Capstone Project' for better visitor experience
- Removed specific week references from project phases for more encouraging presentation
- Maintained detailed project structure while making timeline more flexible
- Ensures consistent 5-star rating for expert-level modules across the framework
- Changed from ambitious app development (computer vision, NLP, etc.) to realistic framework engineering
- New focus areas: performance optimization, algorithm extensions, systems engineering, benchmarking analysis, developer tools
- Projects now align with what students actually built: a complete ML framework
- Emphasizes systems engineering and optimization skills rather than application development
- Maintains 'no PyTorch imports' constraint to prove deep framework understanding
- Added 'Complete System Integration' section emphasizing how all 14 modules connect
- Highlighted that students build ONE cohesive ML framework, not isolated exercises
- Added capstone project section encouraging real applications using only TinyTorch
- Updated README.md 'What You'll Build' to emphasize system integration
- Added visual flow diagram showing module dependencies and connections
- Emphasized 'no PyTorch imports' constraint to prove framework completeness
Key additions:
- og:title, og:description, og:url, og:type, og:image for Open Graph
- twitter:card, twitter:title, twitter:description, twitter:image for Twitter
- Uses astronaut/rocket ship tagline for memorable social sharing
- Proper property/name attributes for platform compatibility
This will enable rich previews when sharing TinyTorch links in Slack, Twitter, etc.
- Added bold formatting to match other modules' style
- Enhanced clarity with more specific descriptors
- Added 'efficiently' and 'with proper broadcasting' for precision
- Now consistent with activations and other modules formatting
- Improves visual hierarchy and readability in built book
Updates the introduction with additional motivational context and a clearer explanation of TinyTorch's purpose.
Emphasizes the hands-on learning approach and the benefits of building ML frameworks from scratch.
Replaces a sentence with an analogy to enhance the message's impact.
- Added 'Prof. Vijay Janapa Reddi (Harvard University)' right after title
- Positioned prominently for proper academic/course attribution
- Matches book config author field for consistency
- Standard practice for educational materials and courses
- Removed redundant 'How This Works' section (covered by Learning Philosophy)
- Removed academic jargon sentence about educational framework
- Cleaned up all em dashes, hyphens, and arrows per user preference
- Changed 'Build → Use → Master' to 'Build, Use, Master'
- Result: Much cleaner, more direct presentation
Key improvements:
- Moved educational framework positioning up front where visitors need it
- Blended 'Science vs Engineering' into more natural 'Core Difference'
- Removed defensive 'Our unique contribution' language
- Changed 'What Makes Different' to conversational 'How This Works'
- Removed bullet points for more natural paragraph flow
- Simplified acknowledgments without academic defensiveness
Result: Much more welcoming and confident presentation
- Added complementary learning reference to mlsysbook.ai
- Positioned as comprehensive systems knowledge companion
- TinyTorch = build systems, ML Systems book = systems context
- Perfect educational pairing for complete ML systems understanding
Updated the course journey section to match the exact navigation structure:
- Foundation: Setup, Tensors, Activations
- Building Blocks: Layers, Networks, CNNs
- Training Systems: DataLoader, Autograd, Optimizers, Training
- Production & Performance: Compression, Kernels, Benchmarking, MLOps
Changes:
- Cleaner bullet format with • separators
- Concise descriptions for each section
- Exact alignment with site navigation
- More scannable and consistent layout
Result: Perfect consistency between landing page and navigation structure.
Changes:
- Replaced em dashes (—) with simpler punctuation
- Used colons (:) for explanatory clauses
- Used periods (.) for sentence breaks
- Removed unnecessary punctuation complexity
Result: Cleaner, more readable text that flows better without distracting typography.
Changed main tagline from:
'Most ML education teaches you to use frameworks. TinyTorch teaches you to understand them.'
To:
'Most ML education teaches you to use frameworks. TinyTorch teaches you to build them.'
Rationale:
- 'Understand' is vague and passive
- 'Build' is concrete and action-oriented
- Aligns perfectly with engineering focus we just established
- Reinforces the hands-on, construction-based learning approach
- More compelling for engineering-minded learners
Updated in both README.md and book/intro.md for consistency.
Key improvement:
- Replaced 'Learning Opportunity' with 'Science vs Engineering' framing
- Clearly positions TinyTorch as ML engineering education vs traditional ML science
- Uses ⚖️ emoji to reinforce the comparison concept
- Bold formatting on key terms: **science** vs **engineering**
- Creates stronger identity formation: 'I want to be an ML engineer'
- Differentiates from theory-heavy courses with concrete value proposition
Result: Transforms value prop from 'better learning' to 'different career path' - much more compelling positioning for engineering-minded learners.
Key improvements:
- Added 'Learning Opportunity' section with positive framing
- Expanded 'What Makes TinyTorch Different' with concrete examples
- Enhanced learning philosophy with complete example cycle
- Moved CTA section lower after building value and understanding
- Added more substance to each section while maintaining scannability
- Improved course journey descriptions with more detail
- Better flow: Hook → Opportunity → Difference → Philosophy → Journey → CTA
- Maintained positive tone without putting other approaches down
Result: More substantial content that builds desire before asking for action.
- Add 'The Big Picture: Why Build from Scratch?' section at top
- Include 'What Makes TinyTorch Different' with 4 key differentiators
- Match the new big-picture-first structure from root README
- Maintain all existing content but improve hierarchy
- Ensure book and README stay consistent
- Move 'The Big Picture: Why Build from Scratch?' to the top
- Add prominent 'What Makes TinyTorch Different' section highlighting unique value
- Emphasize build-first philosophy vs traditional 'use' frameworks approach
- Show concrete code comparison: traditional vs TinyTorch approach
- Better highlight real production skills, progressive mastery, instant feedback
- Reorganize content flow: vision → differentiators → practical details
- Removed 'Home → Module Name' breadcrumbs that added clutter without value
- Chapters now start cleanly with title and difficulty/time badges
- Maintains the useful difficulty stars and time estimates
- Improves visual hierarchy and reduces interface noise
Result: Cleaner, more focused chapter headers
Problem: Grid cards were showing raw HTML code instead of rendering properly
Root cause: README converter was adding new grid cards while preserving
original ones, creating duplicate/conflicting grid sections
Solution:
- Modified book/convert_readmes.py to remove existing grid cards from
source READMEs before adding new interactive elements
- Added regex patterns to clean up grid-related markup
- Regenerated all 14 book chapters with fixed converter
- Grid cards now render properly as interactive buttons
Result: All chapters now have clean, properly formatted grid cards
that render correctly in Jupyter Book
Standardize module endings with motivational section + grid cards:
Added to 4 key modules:
- 01_setup: Foundation workflow mastery message
- 03_activations: Neural networks come alive message
- 06_cnn: Computer vision implementation message
- 09_optimizers: Learning algorithms message
Standard Format:
## 🎉 Ready to Build?
[Module-specific motivational content about what they're building]
Take your time, test thoroughly, and enjoy building something that really works! 🔥
[Grid cards automatically follow via converter]
Progress: 6/14 modules now have consistent endings
- ✅ 01_setup, 02_tensor, 03_activations, 06_cnn, 07_dataloader, 09_optimizers
- 🔄 8 more modules to standardize
Result: Better user experience with consistent motivation + clear next steps
Key Improvements:
1. **Meaningful titles**: Keep 'Module: CNN' format instead of just 'CNN'
2. **Clean breadcrumbs**: 'Home → CNN' instead of 'Home → Module 3: 03 Activations'
3. **Remove duplicate info**: Stop generating redundant Module Info boxes
4. **Use source formatting**: Let READMEs control their own presentation
5. **Enhanced README**: Added Jupyter Book admonition formatting to CNN module info
Results:
- More logical navigation and titles
- Single source of truth for module information
- Better formatted content boxes (CNN example with admonitions)
- Eliminated confusing duplicate content
- Cleaner, more professional presentation
README Updates:
- All modules now use consistent '🔥 Module: [Name]' format
- Removed inconsistent emojis (🧠, 🚀, 📊, 🧱, 🏋️)
- Removed module numbers and descriptive subtitles
- Clean, consistent branding across all 14 modules
Converter Updates:
- Added header cleaning logic to strip module prefixes from chapter titles
- Chapters now show clean names: 'CNN', 'Tensor', 'Setup', etc.
- No emoji or module numbers in final website headers
- Maintains clean, professional appearance
Result: Consistent source files + clean website presentation
- Keep the '🎉 Ready to Build?' motivational section
- Keep the grid cards for Binder/Colab/Source access
- Remove the redundant '🚀 Interactive Learning' section header
- Content now flows naturally: motivation → 'Choose your preferred way...' → grid cards
- Eliminates unnecessary section break and improves reading flow
- All chapters updated with cleaner, more streamlined format
- Updated book/convert_readmes.py to use Jupyter Book grid cards instead of Bootstrap buttons
- Format matches what was in main branch at merge commit 3a687aa
- Three interactive options in clean grid layout:
🚀 Launch Binder - Interactive browser environment
⚡ Open in Colab - GPU access and cloud compute
📖 View Source - Browse Python source code
- Added helpful 'Save Your Progress' tip about Binder sessions
- All chapters regenerated with proper grid card format
- Tested successful book build
