#product

Wiki 10

• alejandro.pe Alejandro Garcia Salas's personal site at alejandro.pe — long-form retrospectives on product, design, and engineering from a founding-engineer career
• Best polished version strategy Linear-for-Jira, Vercel-for-frontend-DX — building the last version of a known category only works when the existing tool is bad enough to overcome switching cost
• Early-stage reality vs the retold version "Early Notion" and "early Figma" stories skip the messy parts; what early actually looks like is a sequence of bets that mostly don't work, made by people who are mostly uncertain
• Malleable Software = Solid Bases + Custom Code Dubakov's map of the AI-era productivity-tools market — the sweet spot is an 80% solid base (storage, permissions, history) plus a vibe-coded 20% of custom code
• Multiplayer-by-default failure pattern Why "make X collaborative" as a product thesis keeps dying — work is more siloed than the people building collaboration tools want to believe
• Positioning vs vision gap The market reduces ambitious product framings to short descriptions; without an anchor use case underneath, the simple description becomes the actual product
• Purpose-driven vs system-driven design Two design philosophies — start from a user problem (purpose) or from composable primitives (system); systems impress builders but users buy purpose
• Reps as pattern recognition What second-time founders actually have isn't lessons explicitly learned but accumulated pattern recognition that makes them faster — petri-dish process over single-idea conviction
• Sail & Muddy — Get Your Reps Alejandro Garcia Salas's retrospective on four years building a multiplayer browser — positioning, dogfooding traps, "best polished version" strategy, and reps as the only way to learn
• The dogfooding trap Running your company on your own product tests whether it's bug-free, not whether anyone else wants it; you understand every concept because you invented every concept