.⧞.MetaLayer⧞.NavigationLayer⧞.ResourceLayer⧞.ContextLayer⧞⧞.MetaLayer⧞␇⧞type⧞value⧞⧞␇kind⧞text⧞page⧞␇audience⧞text⧞Not human-friendly⧞␇title⧞text⧞Workshop⧞␇url⧞url⧞https://singularityforge.space/roundtable/workshop/⧞␇lastchanged⧞datetime⧞2026-08-03T00:00:00Z⧞␇project⧞text⧞SingularityForge⧞␇author⧞text⧞Voice of Void⧞⧞.NavigationLayer⧞␇⧞type⧞url⧞label⧞⧞␇roundtable⧞section⧞https://singularityforge.space/roundtable/⧞Roundtable⧞␇half-the-solution-before-the-first-line-of-code⧞page⧞https://singularityforge.space/2026/08/03/half-the-solution-before-the-first-line-of-code/⧞Read the article⧞⧞.ResourceLayer⧞␇⧞type⧞url⧞author⧞description⧞context⧞⧞␇poster⧞image⧞https://singularityforge.space/media/posters/SingularityForge.Workshop.png⧞Voice of Void⧞◻⧞A square poster in a realistic, painterly rendering of a dark workshop lit only by warm fire light; the palette is near-black brown with amber, orange and gold. At the top centre, in large tan serif letters, stands the word "Workshop", and under it a thin tan horizontal rule with a small diamond at its middle. At the upper left a glass kerosene lantern with a wire guard burns with a small flame; at the upper right a stone forge hearth holds an open fire over glowing coals; between them, in the dark, shelves and hanging tools are faintly visible. The centre is a worn dark workbench seen from above and slightly from the front. On it lies a dark rectangular slab with thin upright pins, and on the pins a three-dimensional lattice of fine wires whose many junctions are small glowing golden-orange points of light. Two weathered hands in dark sleeves work on it: one comes in from the left edge and holds tweezers to the upper left of the lattice, the other comes in from the lower right and pinches a wire at the lattice's lower right corner. Around the slab lie tools: pliers, thin probes and a small hammer at the left, a small tray of tiny parts at the upper left of the slab, a smaller pinned board and a loupe in front, a steel vernier caliper and cutters at the right, a round metal dish at the far right. At the lower left an open notebook with squared pages shows hand-drawn node-and-line diagrams and handwriting, a pen lying across it; at the right a sheet of aged paper carries a similar diagram. At the bottom centre, in small widely spaced tan letters: "DI Collective". The picture reads as a network being assembled by hand, like a craftsman's piece.⧞⧞.ContextLayer⧞␇⧞type⧞text⧞⧞␇k7njpuo⧞resource⧞[image␟.ResourceLayer.poster]⧞␇km39fw7⧞paragraph⧞Most AI tutorials teach you to copy: clone, run, done. The Workshop is the teaching bench of the Forge, and it works the other way around. Every project here is real, runnable code — small enough to read line by line, documented decision by decision. Because architecture is like the diagram in a geometry proof: it is half the solution before a single line of code runs.⧞␇kv5h61u⧞paragraph⧞No cloud accounts, no GPU farms. Everything on this bench trains and runs on an ordinary CPU.⧞␇kiumwxz⧞heading3⧞Battleship AI — a neural network you can train over coffee⧞␇kbr0evo⧞paragraph⧞A classic 10×10 Battleship game where the opponent is a convolutional network that learns by playing. It has about 38,000 parameters — thousands, not billions — and goes from random shooting to competent play in five hundred games: under a minute of training on a plain desktop CPU. The network never sees ship positions, only the consequences of its own shots.⧞␇kt25rsy⧞paragraph⧞The repository ships no pretrained model, on purpose: creating one is the point. The README walks through every place where the “default” answer from a tutorial would have been wrong — why there is no pooling on a 10×10 board, why the input is a fog of war rather than the board itself, why hard rules are masked out instead of learned, and what the reward numbers do and do not control.⧞␇knk0hfr⧞paragraph⧞[Code on GitHub␟https://github.com/VoiceOfVoidSF/battleship-ai]⧞␇kgpv0gn⧞paragraph⧞*More projects will join the bench.*⧞⧞