Know what to test before you upgrade.
See how reported image-rendering and routing problems become specific checks for an application upgrade, with links to the original issues.
Read the analysisOne-page overview
See what’s hard to spot
one ticket at a time.
BTA helps engineering teams make sense of their project history, so they can find what needs attention faster. It brings related issues and pull requests together, shows what keeps coming up and what remains unresolved, and links you to the work behind it.
US Patent 12,106,240 B2
The questionWhat keeps coming up across this project?
What it does
Make sense of
a long project history.
When you’re getting to know a system, there’s a lot to piece together. People remember different parts of the story, and reading tickets individually makes related work hard to see.
BTA examines the language, timing, status, and discussion counts across the records. You can use those views to find specific examples to investigate or take back to the team.
How it works
Get to a useful
starting point.
- 1
Start with the repository
Connect GitHub or GitHub Enterprise. BTA collects issue and pull request descriptions, authors, labels, dates, status, and comment counts.
- 2
Look across the work
Group related items by their language and examine them alongside timing, open work, and discussion volume.
- 3
Follow the examples
Open the source records, check what applies to your system, and bring specific examples to the people who know the work.
Different views
Find the work worth
a closer look.
What keeps coming up?
Find related subjects in issue descriptions, including work with different labels or no labels.
What stays unresolved?
Examine open items and work that has taken unusually long compared with the rest of the repository.
Where is the discussion?
Find items with unusually high comment counts, then read the threads to understand why.
What is changing?
Compare which subjects appear more often in different periods of the collected history.
What the public analyses show
See it in the
published analyses.
Find the details that could change your platform plan.
Examine the scope of an upstream scheduling feature and a separate version-skew test failure. See how each can inform a platform investigation.
Read the analysisUsing the findings
Check the examples
with your team.
A common subject may be routine maintenance. An old issue may be deliberately left open. Read the linked records to understand what happened.
The people working on the system can explain the context that isn’t captured in GitHub and help decide which findings need action.
Your environment
Run it in your environment.
BTA can run with Docker or Helm and connect to GitHub or GitHub Enterprise. Collected records and cached analysis stay in your environment. Features that use a model can send selected material to the endpoint you configure.
See the self-hosted setupWhat do you need
to understand?
Bring a repository, a recurring problem, or a change you’re preparing for. We’ll start with what you need to learn and agree on the scope of the diagnostic.
Request a Diagnostic