Benjamin Coe

my team @sentry just released a feature that lets you create killer dashboards for your app with an AI agent 🤖 it honestly feels a lot like magic -- see my example screenshots. If you use @sentry , try the feature out! feedback is appreciated 👏 github.com/getsentry/sent…

@David28717353 @zeeg @sentry Oh, awesome we're fixing this 👏 Is exporting the whole payload as JSON sufficient for you, we're imagining that the JSON export will include all fields, CSV export will be as it is now (the configured fields).

@BenjaminCoe @zeeg @sentry Sorry, I didn’t explain that well, the logs are actually correct but when I export to CSV I don’t get the full payload, timestamp and message are the only two fields I receive in the export

Folks using our logging product, especially folks who are early stage, what do you wish was easier, or that we'd change?

@David28717353 @zeeg @sentry Hey, Can you give me some examples of attributes that are being dropped. Perhaps by sharing the payload set in Vercel logs and the payload you see in Sentry? We're trying to reproduce and having trouble. Feel free to send me a private message.

@zeeg @sentry I was curious to try it to use as a log drain for Vercel but for some reason you don’t keep all of the data like User Agent etc

@madbyk @sentry not going to lie, the terminal version looks a lot better than I would have guessed very cool.

So I'm starting this friendly rivalary between the CLI team and the Web UI team for @sentry dashboards. Which one looks better?

@tmuxvim @sentry both features are being worked on for metrics 👍 soon.

@BenjaminCoe @sentry LMAO I was hoping someone would say something. can you PLEASE let us use more than one metric series at a time, in dashboards? and do formulas on them (dividing, etc)? thanks for the tz fix, I'll take a look <3

the X axis on @sentry charts makes me want to blow my brains out

used @claudeai to help fix cross-platform tests, getting out a fix for a CVE in a transitive dependency. This would have taken me many hours of work in the past spinning up a VM, etc. github.com/istanbuljs/nyc… the future is cool.

@tolik518 these ReDoS CVEs aren't usually applicable to testing libraries, they're applicable to minimatch being exposed on a user facing route in something like a web application or API. Having said that, go ahead and try: npm i nyc@next

@BenjaminCoe could you please chek out github.com/istanbuljs/nyc… this is a major issue with nyc right now

@zeeg @jreidgreer our own metrics solution is also in beta and has dashboard support: docs.sentry.io/product/explor…

@zeeg @jreidgreer hey @jreidgreer, my team works on the dashboard platform. I would argue it works *particularly well* for span data, as this is where we've done a lot of our development recently. feel free to reach out with any questions 👏

I'm curious how mature the @sentry dashboards solution is. I've been beating my head against a wall trying to get our OTEL + Datadog w/ spans & metrics solution working and thinking how nice it would be to have everything on one platform.

made an extremely niche app with @claudeai this weekend for measuring the angle of adjustable climbing walls... I have to admit, a future where you can whip together something like this *in a morning*, for a very specific use case, is pretty compelling.

working on productionizing your awesome new product? @NikolovLazar and I are putting together a series of posts about using @sentry to identify, and fix, common scaling problems: blog.sentry.io/paginating-lar…

@JoshuaKGoldberg @Sentry welcome 🎉

Job update: I'm going to start as a frontend dev at @Sentry on February 16th! 🥳 Sentry has a stellar product I've enjoyed using, lovely people, and an excellent relationship with open source. I'm really looking forward to joining the team. A few open source FAQs... 1/🧵

Not far enough. IMO, you're ngmi if you sleep.

@NateMatykiewicz @zeeg @nateberkopec @getsentry @dhrumil_parekh Thanks for the details Nate. We're in the process of revamping and improving many of Sentry's dashboards. I'll make sure that Time Spent continues to be used consistently throughout the app, to help with this type of investigation.

@BenjaminCoe @zeeg @nateberkopec @getsentry @dhrumil_parekh The more you can improve the things at the top of that list, the more throughput your app has, and at some point you can probably reduce servers slightly without affecting traffic. Small improvements on heavily requested endpoints/jobs can make huge throughput differences.

In the last ~year, 80% of the clients I get who have ~10 engineers or less have moved to @getsentry for 100% of their observability needs. Perf, errors, logs, and (soon) metrics. Noticeable shift away from Scout, New Relic.

@aarondelasy @zeeg what project stats are most important to you? My thinking is that this part of the problem you're trying to solve can be solved with a gold pre-built dashboard we provide, we can even add stuff that you feel we're currently missing.

oh damn while searching for an answer I discovered that projects are now under "Insights" which makes no sense I guess, but anyway, here's your answer: I used this section of the sidebar to access all my projects, see project stats, access project settings and create new projects. up until this moment I was using this little dropdown and two highlighted buttons. now I guess I will just use "Insights" section, but it wasn't intuitive that projects live there...

give Santa (me) your Christmas wishes for Sentry if you're in this ~10 engineer cohort

@aarondelasy @zeeg this is something on my mind (as a PM at Sentry), what is it that you're accessing the project pages for on a regular basis? I'm curious why you liked those pages being easy to access?

@zeeg bring back sidebar with intuitive way to access your projects

@DouTatsu I'd love to know the sorts of questions you're trying to answer about performance that you're finding challenging in Sentry, we've been working hard to simplify the product surface over the past few months, with more work to come.

@nateberkopec @getsentry Still unable to move from Scout to Sentry - it's too hard to have actionable insights into performance in Sentry, too noisy. I suppose a skill issue on my end, but as long as Scout continues to be so easy to use, unlikely to switch to Sentry for that (errors are all sentry tho)

@NateMatykiewicz @zeeg @nateberkopec @getsentry @dhrumil_parekh I'm curious @NateMatykiewicz, what types of insights are you finding you get from Scout that help you reduce your # of servers?

@zeeg @nateberkopec @getsentry @dhrumil_parekh The 3 questions I ask my APM are: 1. Why is [endpoint/job] slow? 2. What is the slowest [endpoint/job]? 3. What can I improve so that I don’t need as many servers? Sentry answers 1 & 2 well. But #3 keeps me running back to Scout.

Thanking all who came last night's SPDY MEETUP 004, as I had the distinct pleasure of having very timely presentations: 🔸 @RickByers discussed how to ask your boss to work on performance, sharing how he helped launch Core Web Vitals 🔸 @sergical demo'd @Sentry's webvitals.com which is now OSS! A great night . See you all in 2026!