The original idea
Since roughly 2021, Nina has been dreaming about software made for one person, team, or moment, without turning every idea into a conventional software project.
WHY + HOW
WELCOME TO THE MONGODB BUILD LAB(a custom software factory)
A participant enters one prompt. The Build Lab plans, builds, tests, and deploys a web app with MongoDB Atlas and a built-in LLM.
WHY
Close the gap between a prompt and a working app.
A prompt still needs a delivery system
Coding agents can write the app, but intake, planning, safety, database and model access, tests, repair, deployment, and operations still have to work together.
Give the agent a clear path
The Build Lab gives a coding agent a constrained environment, then checks the result before sending the participant a link.
STACK
The pieces behind the factory.
Static web + React
Static HTML, CSS, and JavaScript run the event surfaces. Generated apps use React and Vite.
Node.js services
Node.js runs the APIs, worker, email jobs, generated-app runtime, verification, and publication flow.
MongoDB Atlas + Voyage AI
MongoDB Atlas stores platform and app data and provides MongoDB Atlas Search, MongoDB Atlas Vector Search, hybrid rank fusion, change streams, feedback, voting, telemetry, and judging. Voyage AI provides embeddings and rerank-2.5 for the Build Assistant.
Model gateway, agents, Vercel
Model calls go through MongoDB's internal gateway. Coding-agent adapters build the apps. Vercel and Vercel Blob serve the platform and published artifacts.
LOOSE ARCHITECTURE
One MongoDB Atlas record connects the whole build.
1
Intake and safety checks create the MongoDB Atlas submission record.
2
A planning agent produces a focused brief for attendee review or automatic approval.
3
The worker gives a coding agent an isolated starter app, narrow runtime API, and explicit rules.
4
Source, manifest, browser, and MongoDB Atlas-backed behavior checks drive bounded repair or fallback.
5
The platform creates MongoDB Atlas data and indexes, publishes the artifact, and smoke-checks the hosted app.
6
MongoDB Atlas updates feed tracking, email, Apps, projection, organizer, judging, and feedback surfaces.
Generated browser apps never receive database credentials or model keys. They call a narrow runtime API; the trusted platform validates and performs each MongoDB Atlas or model operation.
BUILD PIPELINE
Plan, build, test, repair, and publish.
1
Normalize the prompt, identity, sharing choice, and event metadata.
2
Run deterministic and model-assisted safety checks.
3
Create and approve a focused build brief.
4
Build from the constrained starter and declared MongoDB Atlas data model.
5
Check source, manifest, requested capabilities, and visible user journey.
6
Run browser tests, including a real MongoDB Atlas-backed mutation.
7
Run a bounded repair pass or smaller functional fallback.
8
Create MongoDB Atlas collections, seed data, and search indexes.
9
Publish and smoke-check the hosted app through the participant path.
10
Update live event state and send tracking and ready emails.
END-TO-END TESTING
The test runner made changes safer and faster.
A real production path
The runner submits an idea, approves the plan, waits for the build, writes a record in the generated app, and checks the public gallery.
Run it after each change
The team can repeat the same path instead of manually tracing failures across services.
Failures become regression tests
The runner keeps diagnostics and turns reproduced failures into tests.
Project scale: 1 sprint · 2 people.
BUILD ASSISTANT