SEV 1 Party

“From Now On” From The Greatest Showman 🎵

In 2023 I wrote down personal tenets (Epic Quests are Born from Clear Intentions) and highly recommend doing this exercise for yourself. It may bring incredible clarity.

But I realized this was missing a HUGE element – I never wrote about engineering principles…or maybe I did? Just a little too much?

—

When you work, people will try to define who they think you are supposed to be as a leader – this is true whether you are an individual contributor or a manager.

But as you continue down your career path, and I know several of my readers are earlier on in their careers, it is ever so important that you define what leadership means for you.

Being a leader isn’t about telling others what to do (a common mistake even senior people will unfortunately model for you…) – it’s about getting their permission to trust you and that you have their best interest in mind. Your peers absolutely should NOT be around you (or anyone) who doesn’t understand that- and your peers can’t know if they can trust you if you don’t say what you mean and do what you say. Sometimes that means making incredibly hard choices.

If you are gonna make a hard choice, might as well write down why before you do. Right? RIGHT?

So in the event anyone is ever truly unsure, what I care about…

Engineering Tenets to Die By

1. I will not fire you for breaking production unless it is malicious intent. I will not assume negligence. And I will not slow you down because you broke it. Tech debt is not individualized negligence – it is systemic and requires partnership along with a quality first mindset. For the monetary and culture data of why, see data in the blogs under “Incidents, Production, & Ops.”
2. I will not let someone else fire you for breaking production.
3. I will advocate for the customer and be a liaison between the above and anything that may happen to build and rebuild trust.
4. I will not hide when production is broken. If I see an instance where transparency has failed, I will self-remediate it to find out what happened and why it was not published. If it happens on the weekend, it will be my top priority to figure out how to bring balance back into your life if you were on-call.
5. I will advocate for reduction or elimination of process that produces undue burden on engineers at the expense of the business instead of in service to it. I will advocate for reduction of process that mistakenly makes us less safe hidden behind a guise of perceived safety. No matter what it takes and even when rampant excuses are used to justify it.
6. I will not let “the way things have always been done” prevent the right thing to happen with regards to cost, security, and shared spaces. For additional thoughts on shared spaces, see “Kubernetes & Multi-Tenancy.”
7. I will make sure your code is not being stalled unnecessarily because of over-engineering and not giving you the chance to try to push to production and learn. Even if it requires a tough conversation that makes smart and kind and well meaning people uncomfortable.
8. I will not green light every idea. I can’t. I will also absolutely slow down a flurry of ideas if I think it may cause unresolved conflict as people “compete” against each other to build the same thing. I will slow it down in an effort to get those interested to work together and partner to speed up. You cannot build a house divided and competing with each other.
9. I will not chase the competitor out of fear. I care about building good products in service to customers based on what they want and what their pain points actually are.
10. I will not race. A desire to race is a sign of business weakness.
11. I will make sure you get credit for your research and galvanize your project. Please give me the time to do so and understand it enough to do so. Executive summaries help me help you.
12. I will not let structure control what needs to get done. I will allow you to cross blurry lines and redraw them. Live outside of boxes in service to the customer.

There have been so many words written on this blog in support of the above. I have done my research – and am committed because of it.

Please do not ever hire me if you think otherwise because you would have to live with the influence that comes with it as long as my butt is at a standing desk and my mouth is on zoom.

I have a reputation – And I’ve been told, I can be noisy which would be really bad for you (but great for your culture) 😘.

It’s too bad there’s so many damn people I can count on and have no actual need for this yet.

So if, instead the above tenets excite the crap out of you and you want to read the data behind why I would be firm on the above with depth and richness and the sassiness that can only be written by a woman in engineering constantly skating the hide-the-pain-harold line that is trying to be both politely assertive and grateful while not offending anyone at all ever, and obviously, failing gracefully – here they are organized:

Incidents, Production, & Ops

SEV 1 Party: Fun posts on CICD, incidents, breaking production, and healthy cultures of failure. A lot of this is supported by Accelerate and Amy Edmondson’s research as well as the history of #hugops.

1. In Defense (Not Really) of Change Management: Hope is a Four Letter Word
2. How Culture Impacts Efficiency: Applying “Accelerate” to Game Studio Culture
3. Cell-Based Architecture: Lower the Blast Radius. Continuous Deployment Is Here.
4. The Invitation: When You Can’t RSVP to the SEV 1 Party Any Faster
5. The Darkest Timeline: When to Host a Failure Party and Nuke All That Process
6. Change Failure by Incident Revenue Loss: What the ’23 DORA State of DevOps Report Misses
7. Yes, I Love Ops: Because We Do Not Fear Production.
8. The Real Way To Pay Down Your RTO: With Kindness, Comes Speed
9. SEV 1 Party: Prod Down. Bridge Up.

Kubernetes & Multi-Tenancy

In case you have too many clusters…

1. In Defense of Less Clusters: Part 3 – The Inverse Conway Maneuver, Team Namespaces, & Business Scalability
2. Cluster Sprawl: How Many K8s Clusters and People To Manage Them Do You Need?: Part 2
3. I Know Places: In Defense of Less Clusters & More Node Isolation: Part 1
4. Self-Service Paradoxes: “Molly! Write about Cluster Vending Machines!”
5. On Hard Problems: Why the CSI Driver for Amazon FSx for OpenZFS Matters
6. Using Self-Managed Amazon Elastic Kubernetes Service Like It’s AWS Fargate
7. Thank You for Playing: How Amazing Teams Scale Kubernetes to Meet the Most Demanding Games
8. The Next 5 years: Streaming Games to Smart TVs backed by Kubernetes
9. Calculating Uptime for a Platform in K8sNaaS and K8sCaaS Business Models
10. Perspective: How Points of View Influence Shared Engineering Conventions
11. “There Aren’t Jr. Eng Roles for K8s” is a Half-Truth Best told Through Uptime

Leadership & Management

1. The Best Leadership Style is the One Where You Don’t Row Alone
2. How to Succeed in A Career Test: Owning Your Leadership Style Under Pressure
3. 6 Techniques to Model Leadership Intentionally And Not Panic
4. Leadership Etiquette: How To Strike By Letting Others Go First
5. One-Shots: When to Roll for Initiative on Your Company Wide Ask
6. Making The Management Game Too Hard
7. The Day I Learned to Break the Rules
8. The Art of the Ask
9. The Best Fake Offsite
10. Annual Performance Reviews: A Quest for Happiness when Comparison is the Thief of Joy
11. Finding Lost Pens & Lending Them: Overcoming the Myths around Growing Scope & Influence
12. Clear Direction that Lacks Opportunity Stems from Confirmation Bias

And remember…epic quests are born from clear intentions.

Header Image by Brett Jordan from Unsplash.