Chasing the Sentry prize for DEV's Summer Bug Smash? Let us know what questions you have.

#bugsmash #devchallenge #performance #monitoring

DEV just kicked off their Big Summer Bug Smash, an event all about fixing bugs and optimizing performance in existing projects.

And guess what? There's a dedicated Sentry prize category. Participants that use Sentry to improve their apps will qualify for Best Use of Sentry. Three winners in the category will receive:

$500 USD
Limited Edition Sentry Skateboard
Exclusive Winner Badge

Live Demo and Q&A

For anyone new to Sentry, we'll be hosting a Live Demo and Q&A Session right on the DEV homepage on July 30 at 10am PT / 1pm ET. You'll also be able to view on YouTube:

Notify Me 🔔

We'll answer the questions you leave on this post during live during our stream.

Have questions already? Drop them in the comments! Want to play around with Sentry first? Bookmark this post!

We can't wait to see your submissions roll in!

Top comments (10)

Ben Halpern • Jul 28

What are the most interesting ways bug detection is evolving? What novel patterns do Sentry's tools support or rely on?

FrancisTRᴅᴇᴠ (っ◔◡◔)っ • Jul 14

Can't wait to see the use of Sentry for this challenge :D

Vinicius Pereira • Jul 14

Does the "Best Use of Sentry" category reward catching a bug that never throws? My strongest use isn't an exception, it's tracing a silent correctness bug in a RAG pipeline: retrieved context gets truncated over the token budget, no error, green tests, wrong answer only in production. Do you weigh surfacing a non-crashing defect through spans/tracing as strongly as fixing a crash or a slow endpoint?

Jess Lee • Jul 15

@vinimabreu this definitely counts and we'd love for you to expand on the details in your submission, if you decide to participate!

Vinicius Pereira • Jul 15

Thanks Jess, already in! The submission went live yesterday: dev.to/vinimabreu/clear-the-lineup...

It walks the full trace: the span where retrieved context crossed the token budget, the silent truncation, and the wrong answer it produced downstream, then the fix verified against the same trace. Glad to hear the category treats silent correctness bugs with the same weight as crashes.

Winter • Jul 29

From my understanding Sentry automatically detects errors when an application crashes and groups similar occurrences into a single "issue" rather than overwhelming developers with duplicate alerts.

Could you walk me through, step by step, what happens from the moment a user's application crashes to when a developer sees a single, organized issue in their dashboard?
Where does AI fit into that pipeline? Does it replace the original matching logic entirely, or does it operate alongside it as a complementary, more intelligent layer?