Use case · Engineering

Find the pattern. In your repo.

The codebase, the API docs, and the engineering wiki sit in one project. A new teammate asks where auth lives. The answer is the file path and the doc, not a generic snippet from the public internet.

50% faster onboardingFile paths, not guessesYour repo and your docs

Play the screens Book a demo

EazyAI · Developer AssistanceSample

service-api · docs · ADRs indexed

Branch main · read only

Where do we check the API key on public routes?

Public routes call requireKey in middleware/auth.ts before the handler. ADR-014 says browser clients use a publishable key, server clients use the secret. auth.tsADR-014

docs/auth.md still mentions a header named X-Token. The code reads X-Api-Key. Both are listed. The code is the one that runs.

What changes on this desk.

The screens above are a sample of the workspace. They play on their own. Select a step to hold it.

Onboarding is a tour of Slack.

New engineers ask where the middleware is, which client to call, and which ADR still applies. The wiki is stale in one place and right in another. Search returns a filename, not the reason.

The file path comes with the explanation.

EazyAI answers from the repositories and docs you connect. The reply names the path and the ADR. When the wiki and the code disagree, both are shown.

How EazyAI is used here.

Four moves. The sources are yours. The answer names them. A person keeps the decision that matters.

01Connect the repo and the docs

The code and the ADRs stay on your network. The agent reads. It does not push.

02Ask like a teammate

Where, why, and which client. Plain language is enough.

03Open the path

The answer names the file and the doc that explains it.

04See the stale page

If the wiki and the code differ, both show up, with the code marked as what runs.

What you can show a buyer.

The working shape of this use case. Live in two to three days on the sources you already have.

Your code, not a sample from the internet

Answers are grounded in the repositories you connect.

A path a new hire can open

Onboarding questions resolve to a file, a line, and an ADR.

Stale docs stay visible

Disagreement between the wiki and the code is shown, not smoothed over.

50%Faster onboarding to the first real change

PathFile and doc on the answer

ReadThe agent does not push code

The file stays.

The project runs on your premises or your private cloud. Dual RAG cites the passage. The model you serve is the one you register. A demo is the way onto a first workflow.

Operating proof for engineering onboarding.

Answers from your codebase and internal docs — patterns, not public-internet guesses.

RepoRepo-grounded Q&A

OnboardPacks for new engineers

PrivateSecure private index

The snippet, the doc, and the file — across your repo.

New hires stop hunting Slack for “how we do auth.” Answers cite the file. Secrets stay out of the index you configure.

Repo-grounded Q&A

Ask how a service works. Get the file and the pattern.

Onboarding packs

Short paths through the architecture you already wrote.

Secure private index

Only the repos and docs you allow.

Before you book.

What engineering leadership asks before a first repo index goes live.

Does it train on our code?

No. Retrieval runs on the index you host. Nothing is used to train a public model.

Can we exclude secrets?

Yes. You choose paths and allowlists before indexing.

Who merges the change?

A person. The agent answers and points at files; it does not ship without review.

Other desks.

Same platform. A different set of sources and a different person in the loop.

See this use case on your files.

Book a demo. A first workflow in two to three days. Your data stays where it is.

Book a demo