Machine learning systems
Attention Is All You Need
Interactive deep dive
A visual walkthrough of Transformer architecture, self-attention, multi-head attention, positional encoding, and encoder/decoder flow.
Read article> deep-dive node online
Systems notes, experiments, and technical findings.
Machine learning systems
Interactive deep dive
A visual walkthrough of Transformer architecture, self-attention, multi-head attention, positional encoding, and encoder/decoder flow.
Read articleLinux memory
Interactive deep dive
A hands-on look at file-backed mappings, MAP_PRIVATE, MAP_SHARED, page faults, page cache, and /proc maps.
Read articleLinux memory
Interactive deep dive
A visual explanation of fork, shared pages, write faults, private copies, RSS, and PSS.
Read articleData structures
Interactive deep dive
A hands-on explanation of red-black tree rules, insertion repair, recoloring, rotations, and logarithmic search.
Read articleData structures
4 min read
A systems-level look at intrusive lists, embedded nodes, pointer chasing, and cache locality.
Read articleLinux memory
Jan 31, 2026 / 9 min read
A glibc allocator walkthrough covering chunk headers, tcache, bins, coalescing, libc leaks, and the kernel slow path.
Read articleCPU architecture
6 min read
A practical explanation of cache hierarchy, cache lines, false sharing, and padding using a multithreaded C experiment.
Read articleLinux memory
Feb 1, 2026 / 13 min read
A practical explanation of demand paging using a restaurant analogy, page faults, page frames, and a Linux malloc experiment.
Read article