Mandeep Singh·Aug 10May I Have Your Attention?Standard attention spends most of its life stuck in memory traffic on “the slow road” between lightning-fast SRAM (19 TB/s) and off-chip…
Mandeep Singh·May 5How DeepSeek V4 Makes Million-Token Context Actually Cheap?Every frontier model claims a million-token context window now. Most of them can’t actually use it. You paste a long document, the KV cache…
Mandeep Singh·Apr 29The AI Layoff Trap: Why Every Firm Acts Rationally and Everyone LosesBack in 1914, Henry Ford baffled his peers by doing the unthinkable: he doubled his factory workers’ pay to $5 a day. On paper, it was a…
Mandeep Singh·Mar 23Beyond Pixels: Meta’s I-JEPA and the Dawn of AI World ModelsIntroductionA response icon1A response icon1
Mandeep Singh·Jan 22Understanding Neural Network Optimizers: A Visual JourneyOptimizers are the engines that power neural network training. They determine how your model learns from data by updating weights to…
Mandeep Singh·Dec 27, 2025From P-Values to Probability: A Better Framework for A/B TestingIntroduction
Mandeep Singh·Dec 14, 2025Gated Attention: Solving the Hidden Bottlenecks in Transformer Attention1) Introduction