Lab
Products we build and run
Alongside client work, we build our own AI products. They run on the same stack and to the same standards, and they show us what breaks in production before it breaks for a client.
Product Launched March 2025
git1file
A tool that turns a GitHub repository into one compact, structured file that a large language model can read in a single pass.
- Compression
- At least 25% fewer tokens than other tools for the same repository.
- Automatic exclusion
- Binaries, logs, lock files and redundant documentation are filtered out.
- XML output
- A structured format that keeps the file tree and the code relationships clear to a model.
- Private repositories
- Secure access to private code through the GitHub integration.
acme-api.xml Example output
<repository name="acme/api" files="38">
<tree>src/ server.ts routes/ lib/ tests/</tree>
<file path="src/server.ts">
import { createServer } from "./lib/http";
…
</file>
<file path="src/routes/invoices.ts">
…
</file>
<excluded reason="binary, lockfile, generated" />
</repository>Also in the lab
Open work and programs
- Open source
Nina Labs on GitHub
We publish tools, examples and experiments from our own work on GitHub.
- Program
Microsoft for Startups
A member since January 2025, with Azure credits, technical guidance and access to Microsoft's startup network.
Want an AI product of your own?
We build internal tools and customer-facing AI products with the same team and the same standards as our own.