Ian Fahey, Analytics Engineer at dbt Labs

Hey there 👋 #

Hey folks, I’m Ian (he/him) and I’m a new analytics engineer on the data team at dbt Labs (I’m currently writing this in my week 4 (!!) at dbt Labs). I’m super excited to talk about walking through my onboarding experience, working through imposter syndrome, and adjusting to a new team.

Before we dive into it, some things to know about me:

• I’m part of the very common (cough) actuary to analytics engineering pipeline; I spent the first 12 years of my career as an actuary and have always loved numbers and data, but escaped the mortality tables into Tableau and eventually business intelligence before finding my calling as a fully-fledged purple person and analytics engineer.
• Ball is life D&D is life: I’ve tried a lot of hobbies, but Dungeons and Dragons is it for me. I’m currently playing or running three different games now and dedicate several hours of my week to prepping or reading up on D&D.
• I’m most energized by getting stakeholders the data they need; when someone gets unblocked by a data pipeline I work on, that’s what’s really rewarding to me and a big reason why I like to be on a data team.
• Ball is life, D&D is life Documentation is life: I’m a big believer in answering a question once, so I try to document everything, from dbt models to data analysis work, as thoroughly as I can. I used to be a data person who ran down a lot of fly balls—someone who was doing a lot of data work and being helpful, but not democratizing that work in a way that was sustainable or scalable. Remembering those early days, I’m now pretty dedicated to writing meaningful and helpful documentation when and wherever I can.

Orchestrating my day #

8:32 a.m. My mornings start with some carefully curated vibes and a whole lot of reading. Since my team is mountain and west coast leaning, my mornings are naturally a little quieter until the entire team is on Slack. As a result, I spend my mornings doing what I call my “homework time” where I catch up on any important readings or write up any important documents.

Some of the things I’m reading during my onboarding process include:

• How we structure our dbt projects guide
• dbt Labs’ style guide
• My 30/60/90 document
• Our internal dbt project’s code, documentation, and DAG
• Ongoing project updates
• General dbt Labs onboarding materials (HR, IT, org stuff)

This is a pretty long and important list, which is why it’s nice to have some dedicated time in the morning to get through it all.

Beyond enjoying reading in a calm environment, my mornings are also probably my favorite time of the day because of Slack. This is a red hot take, but I actually like Slack. As an extrovert and former AIM kid, I love perusing through different Slack channels, getting involved in threads, and chatting with folks, especially people I don’t work with on a regular basis. Since dbt Labs is a remote-first company, my mornings where I spend some time in Slack make me feel like I’m in somewhat of an office and provide a semblance of a community to me.

11:00 a.m. Around late morning I’ll start to check-in with folks I’m working with things on or partake in a standup. Since I’m in the process of onboarding, members of the data team have been walking me through existing work to help me get more dialed in on the state of affairs for the data team.

For instance, Andrew and Brandon talked through the various ways we can pull granular user activity into the warehouse (without accepting the entire firehose). Just listening in on an explanation of these problems and how the team thinks about them helps get the juices flowing for further partnering throughout the day.

1:15 p.m. Since I’m an analytics engineer embedded on the product team, part of my 30/60/90 goals is to meet with our product managers to learn more about them and their work. So far, I’ve had the pleasure of chatting with Jeremy Cohen, the product manager for dbt Core, Julia Schottenstein, the product manager of shipments, and every PM in between. Today, I caught up with Margaret Francis, our Chief Product Officer, and we transitioned from board game collections to Single Tenant considerations over the course of thirty minutes.

2:24 p.m. Andrew pinged the data team that he was deeply invested in optimizing a query that is a considerable drain on our incremental dbt Cloud job and its underlying Snowflake usage. Ric and I jumped onto a video call with him to debug and work through possible solutions. In this case, the ephemeral model crafted boolean flags out of two different partitions, so we discussed refactoring, rematerializing, and Snowflake clustering, none of which improved performance in small-scale testing. We parted ways without a solution.

After our video call, Andrew wrote up what was essentially a “How Not to Make a Lightbulb” document that showed everything we tried and why it didn’t work. He opened up that document to our entire organization to get feedback and soon folks were jumping right in to offer additional ideas.

There are a lot of very smart and thoughtful people who work at dbt Labs, which meant my imposter syndrome came back in full force when I started in this new position. That said, we all have those “head banging against the wall” moments, and it was valuable for me to see other folks feel comfortable raising their hands to be like, “Hey, we bounced a long-running query off of some familiar troubleshooting solutions and came up with nothing.” For me, overcoming imposter syndrome stems from contributing to work and creating value for the people I work with, but it can also come from the courage to ask for help. Andrew’s write-up showed me how people at dbt Labs embrace vulnerability in order to fully contribute to the knowledge loop, both of which are things I look forward to doing as I grow into my role.

5:38 p.m. The day is winding down, which means it’s time to write up some “love letters” to myself for tomorrow. I said it earlier, but I love documenting my work. My initial docs are almost always selfish, written for an audience of my future self in a moment of maddening forgetfulness. I, therefore, try to treat any documentation I’m writing as a love letter to myself—a meaningful, supportive breakdown that includes everything, even the easiest bits that are so top of mind, I can’t imagine forgetting them.

I’ll often Slack myself in the evening with messages for the following morning summarizing what’s been done and what’s left to do, so I can have a clear understanding of my priorities for that day. Then, I set a 9AM Slack reminder for that message (alongside the many, many others I pulled from various channels throughout the day) so I can step away.

Because I’m working in an atmosphere this motivating and interesting, it’s also incredibly important that I dedicate explicit time to be truly logged off to avoid burnout. After I sign off, I’ll usually unwind by petting my dog, eating dinner, getting some household chores knocked out, and catching up with my wife.

Looking toward the future #

One of the hardest things about leaving my old job and starting a new position here was leaving my data team behind. I really cared about the team and our data stakeholders at my previous job, so I warned Ric that my excitement to join would also come with some grieving.

I’m really grateful that the data team I joined at dbt Labs immediately welcomed me as a part of the team by suggesting slack channels, directing me to documentation, and just goofing around on Slack between action items. Working on a team that cares about me and each other has made the mourning period a little less painful, and definitely makes me hopeful for the future :)

Speaking of the wildly daunting and exciting future, there’s a lot to look forward to as I ramp up my onboarding. I’m probably most excited to explore the boundary between all the good, thoughtful practices at dbt Labs and the skills, experiences, and knowledge I bring to the table. I’ve thought a lot about what good data modeling looks like and so have folks at dbt Labs and in the dbt Community, so I’m interested in learning how I can marry those things together and contribute to the knowledge loop myself.

The first time I made a comment that Ric incorporated into our entity resolution work, I legitimately ran upstairs and told my wife, “I contributed!! I changed a field name!!” Small though it was, it felt like work, like impact, like participation in the vocation I chose. Data work at dbt Labs isn’t done yet (and is it ever really done?), which gives me ample opportunity to learn, grow, and contribute to this team and our stakeholders. Here’s to many more “running up the stairs and celebrating with/at my wife” moments.

Last modified on: