GhostStack
Serverless productivity stack that converts work-call notes into structured end-of-day reports.
An engineering studio architecting full-stack platforms and applied data systems. Currently building GhostStack — a serverless productivity stack that converts ad-hoc work-call notes into structured end-of-day reports.
Serverless productivity stack that converts work-call notes into structured end-of-day reports.
LinkedIn meets Houzz for skilled tradespeople — identity, portfolio, and marketplace in one platform.
We don't assemble systems from defaults. We architect domain-specific platforms where the data model, API surface, and user experience are designed together — not patched together.
React Native and Expo applications built with offline-first sync, type-safe APIs, and production-grade architecture from day one.
NestJS and GraphQL services with PostgreSQL, code-first schemas, JWT authentication, and end-to-end validation through Zod.
Relational schema design and normalization built from first principles. The data model is the product, not an afterthought.
Graduate-level data science applied to real product surfaces — inference, modeling, and decision support beyond vanity dashboards.
Most engineering teams pay an invisible tax on integration debt — the friction between mismatched data models, hand-rolled validation, and contracts that drift over time.
We design PostgreSQL schemas, GraphQL and REST contracts, validation layers, and mobile clients as one continuous system. Drizzle, Zod, NestJS, and React Native, wired together so the types flow end to end.
Base Layer Labs is the engineering studio of Dave Wilson — a full-stack developer pursuing an MS in Data Science. The work spans mobile platforms, backend services, and data systems, with a focus on production-grade architecture and first-principles thinking.
Current focus: GhostStack — a serverless productivity stack that captures ad-hoc notes from work calls and converts them into structured end-of-day reports. Parallel work on TradeFolio (a React Native and NestJS identity layer for skilled tradespeople) and ongoing study in applied statistics and relational data modeling extend the engineering surface area.
Strict correctness, type-safe contracts, and verifiable behavior over rapid, fragile output.
Every system starts with the data model. Schemas, validation, and user surfaces evolve together — never separately.
Whether you're interested in collaboration, acquiring one of our products, or have a question — the line is open.
Remote — United States
Available globally
We'll review your inquiry and respond shortly.