The software decision engine
Build. Buy.
Bridge. Beware.
Software categories scored on five dimensions, with a batch reviewed each night across the quarterly scoring cycle. Independent and evidence-graded. Built and maintained by an operator and a fleet of AI researchers.
Software decisions for the AI age.
Consider owning it. The category differentiates you competitively, and the evidence makes building a scoped capability realistic. Test the business case and your capacity to run it. Teams should build the software that gives them competitive advantages.
SEE THE BUILD CATEGORIES ↗My build portfolio
Browse all →Life OS
● LIVEA cognitive system that reads my life across three contexts and hands the day back already synthesized.
4,141 ENTITIES · 26 MCP TOOLS · 579 TESTS
B4 Index
● LIVEThe build-vs-buy argument turned into a scored, self-rescoring public index.
1,605 CATEGORIES · 11,031 VENDOR FILES · DAILY
$ route --task extraction
→ tier: mid · gate: fresh-context
✓ done-condition verified
Claude-ops
● MITThe operating layer for frontier models: routing, delegation contracts, review gates.
25 SKILLS · 20+ MCP SERVERS · PUBLIC REPO
Observatory
● LIVEUsage intelligence for Claude Code. Every session traced, costed, and separable from test traffic.
OBSERVABILITY · DEV TOOLING
Compound Publishing
● ACTIVE BUILDAutonomous AI book production and publishing pipeline with human approval gates.
CONTENT PIPELINES · ORCHESTRATION
Video Factory
● LIVEPrompt-driven video on Remotion. Schema-validated compositions rendered to MP4 from JSON.
CONTENT PIPELINES · DEV TOOLING