Building services is important, but what happens after they are built and running in production? How do we establish trust with our customers that our service will actually be available? Who creates these definitions and how do we measure them? Service Level Indicators (SLI), Agreements (SLA), and Objectives (SLO) are central to an operations mindset and foundational tools for effective Site Reliability Engineering. This talk will take you on a journey through Springfield as we discuss exactly what SLIs, SLAs, and SLOs are, how to measure them, what targets should be measured, how to define uptime, availability, and acceptable error rates, and what happens when they are breached. Attendees will leave with a clear understanding of how to monitor and report for their services, how SLIs, SLAs and SLOs can aid in this process, and how to implement them within their own teams.

Katie Simmons, a data engineer at ActiveCampaign, will speak about the challenges and benefits of using Airflow for ETL at a rapidly growing company. ActiveCampaign has many thousands of databases - some including tables with up to a trillion rows - several APIs and new source requests coming in every week. This lightning talk will be an overview of using Airflow to extract, load and transform that data into our data lake so that it can be used for Business Intelligence and Data Science.

I first encountered SQLAlchemy several years ago. I didn't get it. It seemed every line I attempted to write would drop me into 50 tabs of labarynthine documentation. Why do we have the ORM *and* Core? Should I build my tables as `Table` instances or should I be extending `Base`? How is `Base` more declarative than a function that returns `Table`s?? Can I please just write SQL??? :sob: I'm still hesitant to peek too far behind the curtain, but I do think I've finally wrapped my head around the philosophical underpinnings of the library and the different problems SQLAlchemy allows us to solve. After all, who among us works with databases that aren't problems in and of themselves?

Not all data is easily accessible. Taking info from a website that requires authentication, interaction, or even just to load a fancy script. This talk will discuss using Selenium to level up your web scraping skills, with examples and suggested practices.

Failure can be scary. There are real costs to a company and its users when software crashes, models are inaccurate, or when systems go down. The emotional stakes feel high-- no one wants to be responsible for a failure. We can lower the stakes by creating spaces to learn from failures, and minimize their impact. This talk introduces two ways to address failure: blameless post-mortems, to learn from an incident; and pre-mortems, to identify modes of failure upfront.