AI coding at home without going broke

the hardware you buy today may look like a bad bet in a year. for most people, renting open source models from a provider at API rates is the right call. a blend of frontier subscriptions for hard thinking and API rates for open models—lean on spec driven development so the expensive models produce the plan and the cheap ones fill it in.

practical breakdown of the three approaches to local AI development without handwaving. the author’s actual take matters: self-hosting only wins if you can keep hardware busy (most can’t), API rental avoids the sunk cost trap of fast-moving hardware, and frontier subscriptions hit hard ceilings on token volume. the hybrid strategy is the useful insight—use expensive models for planning, cheap ones for execution. echoes older patterns from distributed systems (don’t buy the tier-1 hardware, rent commodity). numbers given ($1k to ship what 20 engineers produce in a month) are directional, not gospel, but usefully concrete.