Entity Builders Product Engineering System
A monorepo and spec-driven workflow for validating multiple products without rebuilding the same foundations every time.
Independent product work can collapse into scattered prototypes. Entity Builders is the counter-system: one workspace, shared rules, reusable packages, app-scoped context, and durable OpenSpec changes that let ideas move from fuzzy to shippable.
- Yarn workspace monorepo
- Shared platform rules
- OpenSpec product memory
- Multiple active apps
A solo builder needs speed, but speed without structure creates repeated work and forgotten decisions. The challenge is keeping validation fast while preserving enough architecture and product memory to keep compounding.
- Use web/PWA-first validation unless a native platform reason is documented.
- Keep apps thin and move reusable business logic, AI orchestration, analytics, and infrastructure helpers into packages.
- Use OpenSpec for durable product memory before substantial behavior changes.
- Treat specs as working tools, not bureaucracy, with progressive rigor based on risk.
Workspace architecture
- Multiple products under `apps/` with shared packages and backend infrastructure.
- Reusable patterns for Supabase, AI, analytics, deployment, and app-scoped agent context.
- Portfolio, Tablia, Flowtranslate, PostalPeek, and other products sharing one operating base.
Product memory loop
- OpenSpec proposals, designs, task lists, and permanent specs.
- Agent rules that preserve architecture, database, platform, and workflow decisions.
- Verification scaled by risk so small changes stay lightweight and risky ones get stronger checks.
- A shared system adds ceremony, so the process must stay lightweight enough for a solo builder.
- Reuse is valuable only when it does not slow down validation.
- AI-assisted work needs durable context or every session starts from scratch.
Founder and solo product engineer designing the monorepo, operating rules, app patterns, OpenSpec workflow, and reusable product infrastructure.
- Juan thinks beyond tickets and can design an execution system around ambiguous product work.
- He has taste for operational leverage: reusable code, durable decisions, and fast validation loops.
- He can connect product strategy, architecture, AI-assisted workflows, and delivery habits.