Celebrating 2y as a Chronaut
In the summer of 2022 I started working for Chronosphere as a developer advocate
And today, have reached my 2 year Chrono-versary which is a major personal milestone!
This is typically about when the honeymoon period with a company ends for me and if I don't agree with the strategic bets being made or see evidence that persistent problems are not actively acknowledged and addressed I bounce.
Like I haven't hit a full 2 years anywhere since my first ever tech job at New Relic many many moons ago. This is a big deal for me and while 1-2y short stints aren't uncommon for the slice of b2b SaaS dev tooling startups I’ve been working for... I’ll let this Kris Nóva quote speak for me
Unfortunately due to a severe case of burnout my engineering career came to a screeching halt and I had to accept that any flavor of SWE/SRE was off the table for good. Finding a new career was not something I had on my 2022 bingo card but I figured the easiest transition would be to leverage the skills I'm strongest in (public speaking, training and enablement, having Opinions about observability) while sidestepping my weaknesses (selling things, accounting, waking up early) and stick to a familiar domain. So I was cautiously optimistic when a friend sent me the dev advocacy opening at Chronosphere...
Why I (Still) Choose Chronosphere 2 Years Later
The People, Duh!
It sounds trite but the first thing that keeps me here is the people! There is no shortage of observability expertise and I'm constantly learning new things from fellow Chronauts. One of my fave channels,#prom-queens, is our bat signal for getting help debugging gnarly PromQL queries and has totally leveled my skills up.
The open atmosphere for learning and sharing knowledge is a big win and made possible by hiring folks who aren't just smart but also kind. I have yet to run into a "brilliant jerk" 😄
Those Bennies
(bennies = benefits)
Before Chrono I would've been the first to say that unlimited PTO is a big scam that saves companies money and I still think ymmv depending on your role but my perspective definitely shifted after this past year.
During the icepocalypse in PDX my house lost power plus the water pipes burst all while the roads were encased in a slick sheet of ice and next to my house giant trees terrifyingly swayed and branches creaked under the load of snow. My family and I ended up staying at a hotel for weeks. Without unlimited PTO I would've had to either eat into vacay days (which how horrible as this was anything but a relaxing experience) or "pretend" to work by logging on and trying to block out increasingly paranoid thoughts that at any moment a tree could fall on my house and ruin very sentimental irreplaceable things. I really appreciated requesting those weeks off and knowing that I could still take actually relaxing PTO during the rest of the year.
Also I love our Forma budget which has covered all my nice soft and sustainable merino wear from icebreaker and smartwool and the cutest workout gear from popflex active. If you haven't tapped into your learning/development or health budget this is your sign - use it or lose it!
Oh! and bonuses! Idk what it was about SRE but I can count on half a hand how many c-a-s-h bonuses I got before Chronosphere - I am a big fan because I really don't put much stock in options (pun intended).
The Platform I Wish I Had As An SRE
The bulk of my focus is on the open source side of the house, helping folks learn and get started instrumenting with OpenTelemetry and friends but I personally wouldn't be comfortable being "the face" of Chronosphere if I didn't actually believe in the platform. So forgive the shameless plug but I want to share just 2 of my notable faves:
- I am o-b-s-e-s-s-e-d with our Behavior driven trace sampling that doesn't require a PhD in tracing to configure. Tracing totally gets a bad rep for the steep learning curve and I'm really stoked how we tackled this, you can get value on day 1 just with Baseline. Instead of wrangling thousands of fine-grained tail sampling rules we let you describe traces as slow/fast, large/small and failed and categorize any of those facets as "high value" or "low value". This is sampling designed for humans not the 1-2 tracing wizards in your org. Love it!
2. OK and Lens? Sure looks like a regular degular APM view until you realize that we've constructed this entirely from open source instrumentation across different ecosystems and NOT from our own bespoke agents and telemetry schema. Got Prom metrics, OTel traces, and a mess of logs? Lens not only unifies this but surfaces helpful insights and I think that's cool as hell!
This barely scratches the surface of what we've been building and I'm not at liberty to disclose the awesome stuff in progress but let's just say I am excited by what's to come...
No Predatory Pricing
Part of the reason I worked for so many o11y vendors is a sense of job/industry security - getting visibility into wtf is happening with your systems is essential, not a "nice to have". But I find it kinda gross that lock-in with proprietary agents has allowed vendors to run amok with pricing.
As someone who has experienced multiple sprints being upended to deal with surprise observability overages this is a big one for me. There are DAYS of my life I will never get back spent attempting to square the numbers on a complex billing invoice with a “helpful” usage dashboard and squinting to figure out what data could safely get dropped to get under budget. What took me eons in a previous life would’ve taken just minutes with the Control Plane. Providing a crystal clear view into your telemetry usage is major but the cherry on top is actually giving you the tools needed to quickly and confidently shape, control and manage telemetry volumes.
Because of the staggering volume and cost reduction our customers get with the Control Plane sometimes Chronosphere is painted as "oh they're cheap" like we're peddling fake designer bags or something. It's really more that with us you're only paying for data that actually matters to your business. Instead of encouraging you to "just send everything! all telemetry is totally equally valuable, trust us 😉" and profiting from overages or cardinality spikes we'd rather have you set up team or service based quotas and proactively prevent nasty billing surprises. (side note: cardinality isn't bad in and of itself, you just need to spend it where it counts!)
Lock-In? Not here!
We're built on open instrumentation standards and with the rise of interoperability between projects like Prometheus, OpenTelemetry, Fluent Bit this means "instrument once, observe anywhere" is becoming a reality.
The benefits of owning your telemetry from instrumentation and the pipeline from source to destination are huge. When rerouting telemetry from one platform to another only takes a few config tweaks and a deploy instead of an entire re-instrumentation migration initiative..... that's some leverage!
I believe we're still in the middle of this tectonic shift (remember how long the digital transformation and moving to "the cloud" took? it was many many many years). Obviously I'd love it if you were to come observe your system with Chronosphere but ultimately I think it's a win for us all that as open instrumentation standards mature and become ubiquitous vendors have to start competing on value and functionality instead of relying on lock-in.
Also! This means the content from me and my team (or my pals at competitors) can help you even without being a prospect or customer. My team's O11y workshops are not gated, don't rely on signing up for any account, and gets you started with Prom, Fluent Bit, Perses and OTel. That's the best part of the job imo, getting to grow and invest in the open source community <3.
Leadership that Gets It
ngl I was wary of joining yet another startup founded by engineers having been burned in the past by assuming that infra, operations and reliability would be on equal footing with application and software development.
and I was happy to learn that Martin and Rob really do get the importance of the ops and infra side of the house having been there themselves.
also they introduced me to Tim Tams and my life is 1000% better because of it
as the SRE saying goes "reliability is the most important feature" but this is especially true for an observability platform and my fears were eased when I saw that Reliability & Responsibility was literally one of the company values
We value this so much there's actually been a couple times I thought to myself "hmmm maybe I could do SRE again if it was in an org like this"... but this is just wishful thinking as I am off-call for good 😅...I mean I even named my podcast Off-Call so I'm pretty committed at this point lol.
Best Of Highlights
I have done some of the best work in my career at Chronosphere and here's the shortlist of things I'm most proud of:
- Soft launched Off-Call the podcast where we meet the people behind the pagers (Official™ launch to come next week)
- Developed and lead an Intro To Instrumentation OTel workshop at KubeCon last year
- Presented at my all-time fave conferences (SRECon & Monitorama)
- Wrote fun blogs like Is Your Incident Response Better Than An Energy Company's? & What's the Biggest Unsolved Problem within Site Reliability Engineering?
Closing Thoughts
No company is perfect but it is notable in my book that I've made it to 2 years at Chronosphere and am looking forward to many more.
If you're interested in working together you can check out our open roles here https://chronosphere.io/careers/ and if we've worked together in the past, hit me up for a referral link 😄
Or if you're like "damn I really need Chronosphere" check out some self-guided interactive demos or get in touch with sales .
CAT TAX
Member discussion