Software Bits Newsletter
Subscribe
Sign in
Home
Archive
About
Involutions: The License to Update in O(1)
Why Zobrist hashing updates a chessboard in constant time, and the property that makes incremental computation possible.
Jan 14
•
Taras Tsugrii
4
1
Latest
Top
Discussions
Sparsity: The License to Skip
Why ignoring most of your neural network is the key to efficiency
Jan 7
•
Taras Tsugrii
4
Domain Transformations: The Art of Finding Easier Spaces
Why logarithms prevent underflow, why Fourier speeds up convolutions, and how choosing the right space makes hard problems tractable
Jan 5
•
Taras Tsugrii
4
Beyond Snowflake: A Problem-Solver’s Guide to Distributed ID Generation
How to discover solutions, not just memorize them.
Jan 4
•
Taras Tsugrii
3
Linearity: Why Batching Works
And the property that makes neural network training computationally tractable
Jan 3
•
Taras Tsugrii
4
1
Commutativity: Why Transformers Need Positional Encodings
And other consequences of order not mattering
Jan 1
•
Taras Tsugrii
2
The One Property That Makes FlashAttention Possible
What computing an average teaches us about the most important optimization in modern ML
Dec 29, 2025
•
Taras Tsugrii
6
PyTorch's compilation and matrix multiplication.
Or a case of missed optimizations
Oct 29, 2023
•
Taras Tsugrii
3
See all
Software Bits Newsletter
Short stories on software design and performance.
Subscribe
Software Bits Newsletter
Subscribe
About
Archive
Sitemap
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
