Bugsnag Engineering Leadership Forum San Francisco 2019

For the first time in history, software engineering leaders gathered to discuss the advancement of application stability and share insights on how they align teams across engineering, product, release, and QA to the central stability score.

Speaker videos

Jordan Golinkoff

Senior Data Scientist, Pandora

How we use Bugsnag to prioritize our work
“If you’re getting your quality signals from your app reviews or from emails from your listeners, you have failed.”

Jane Mitchell

Director of Engineering, Within

Making error monitoring actionable with Bugsnag
“By working closely with Bugsnag, utilizing what we have, and what they have to offer, we managed to get to 99.5% stability within the quarter.”

Pierre-Yves Ricau

Android Tech Lead, Square

Hacking Bugsnag for fun and profit
“For every crash, you should ask yourself, ‘Is there any information that I missed?’ ...If there’s information that I wish I had, then I go in the code and add it, and next time I have it.”

James Smith

CEO & Co-founder, Bugsnag

Software Application Health: Evolution and Use Cases
“For complete visibility into the health of web and mobile apps, we’ll need to measure and optimize observability metrics beyond stability with an App Health Scorecard.”

Our guest speakers discuss their approaches to creating debugging workflows, managing stability KPIs, defining team structure and ownership, and implementing error monitoring best practices for the long term.

Panel

Jordan Golinkoff
Jane Mitchell
Pierre-Yves Ricau
Hosted by James Smith

Key learnings

Define ownership and communication work streams

There’s no point finding bugs, unless you have a path to fixing them, and that has to be a team process change. Define ownership and communication work streams for managing application stability.

Crashes reveal that we may need better process

Crashes reveal more than just bugs. They expose a bigger picture operationally — that we may need better process, better ownership, more QA, more collaboration, and better knowledge sharing on systems working together.

Investigate the context of crashes

Keeping a low crash rate requires investment, not only in the pipeline and the tooling, but also in the investigation of the context of crashes.

Segment by app version and key customers

Not all crashes are equal. Prioritize engineering efforts by segmenting error data by app version and key customers.

Make predictive models about where to invest

Look back on error data, and make predictive models about where to invest engineering team's resources.

Correlate user behavior with bug history

Correlate user behavior with bug history, and define your customer experience roadmap by building a rich data record.

Attach diagnostics to the root cause of problems

Several factors can cause a crash. Attach custom metadata to error reports to rule out possible sources. Attach diagnostics to the root cause of problems.

Set stability as a KPI

Set stability as a KPI for engineering teams to improve morale and efficiency.

Tailor Bugsnag for your unique requirements

Push the limits on Bugsnag, and you’ll discover the infinite ways to hack it for your unique requirements and needs.

Attendee takeaways

Brendan Weinstein

CEO, BaseBeta

Matt Robinson

iOS Engineer, Strava

“The benefit of arbitrary metadata and the ability to do things like breadcrumbs for us has been essential."

“It was good to reaffirm the mindset or the mantra that you don’t need to fix every bug.”