[ developerbench.com ]

Find what's failing. File it where you already work.

Dev Bench watches your app for exceptions, failed and slow requests, and the “success” that wasn't: the handled error behind a page that told the user it worked. It triages each problem with your server's own log lines, files it in your GitHub with who was affected, and reopens it when it comes back.

Someone's always watching.

[ rails ] the whole install
bundle add devbench
bin/rails generate devbench
# then set one value in your app's environment:
DEVBENCH_DSN=https://<public>:<secret>@adt-ingest.onrender.com

The generator adds the browser sensor to your layout and your CSP. No sidecar, no browser key, no initializer. Also for Go and any web page.

[ 01 ] what it catches

Everything that went wrong, including what nobody reported

exceptions

Errors, server and browser

Uncaught exceptions, errors your code reports, failed jobs (ActiveJob, Sidekiq), panics, console errors.

failed requests

Every request that failed

Any fetch or XHR that got a 4xx or 5xx, or no answer at all, seen from the user's browser.

slow

Slowness people feel

Slow requests, slow page loads and slow interactions, counted per route rather than per page view.

silent failures

The “success” that wasn't

Your server caught an error and answered 200 anyway. One line marks it, and the browser knows the user was told something worked when it didn't.

[ 02 ] how it works

Sensors → triage → your GitHub

sensors

In your app

The Rails gem, the Go SDK and the browser sensor fingerprint and count problems in-process. Traffic grows with distinct problems, not page views. Messages, stacks and log lines are redacted before they leave your app.

triage

Reads the evidence

Each new problem is triaged with the stack, what the user was doing and the server log lines of that exact request, so the write-up says why, not just where.

your github

One issue per problem

Filed in a repo you choose, with how many users and accounts were affected. Close it when it's fixed; if it comes back, the issue reopens itself.

Support asks “what happened to jane@example.com?” Look her up by email or account and see every problem she hit, with links to the issues.

[ 03 ] vs sentry

Coming from Sentry

What's the same

  • [=]Uncaught and reported exceptions, server and browser
  • [=]Failed jobs, failed requests, slow pages
  • [=]Who was affected, per problem
  • [=]A one-line install

What's more

  • [+]Silent failures: handled errors behind a 200
  • [+]Log-aware triage: the request's own server logs, redacted in your process
  • [+]Issues in your GitHub, not another inbox; regressions reopen

What's not, by design

  • [-]No session replay. We don't record your users' screens.
  • [-]No per-page-view tracing; slowness is counted per route instead.

Run both side by side for a few weeks and compare what each caught.

[ 04 ] editions

Two editions. Billed monthly by invoice.

Sales-led: we talk first, then start a pilot. No card form.

[ V1 ]

Support tool + support developer

$10,000 / month

  • Exceptions, failed and slow requests, and silent failures, server and browser
  • Triage that reads your server logs for the failing request
  • Issues filed in your GitHub with who was affected
  • Regressions reopen themselves
  • “Who was affected” lookup by email or account
  • Fixes, as they land
Talk to us about V1
[ V2 ]

A development team

$20,000 / month

  • Everything in V1, plus:
  • Feature requestsV2 roadmap
  • Ask the AI about a task's statusV2 roadmap
  • A task listV2 roadmap

V2's additions are being built now. We'll tell you plainly what is live when we talk.

Talk to us about V2

See what your app is hiding.

Tell us about your stack. We'll set up a pilot on staging.

Talk to us →