A day in the life of Mapbox Telemetry engineer, Nick Cordella
By: (you guessed it) Nick Cordella
Nick Cordella is an engineer on the telemetry team at Mapbox, where he builds valuable products like traffic predictions from large location datasets. Prior to Mapbox he was a senior data scientist at a startup that built lane-level maps of autonomous vehicle trajectories.
7:00 am
Did I just hear a squeak in the next room or no? I peer at my 5 month-old Charlie’s crib and try to discern if he’s stirring. He seems to be snoozing, but I’m awake now with little chance of falling back asleep. I sure would love to stay for some morning snuggles but it would only make leaving harder, so I grab a piece of fruit and head out the door while my wife, Meghan, and the baby are just rousing. Big shout out to Meghan for handling the morning routine before our nanny arrives!
7:30 am
Like many Oaklanders, I have the commute down to a science. Give me a time of day and I’ll tell you whether your best bet is to take BART, bus, ferry, or casual carpool into and out of the city (San Francisco). Under ideal conditions, I’m leaving the house now or earlier, and the casual carpool line around the corner isn’t too long. This means I’ll be able to have a seat for my commute from which I can fully dive into my laptop and get a start on work. Hopefully, I can zip through all my notifications from Slack and Github on the car ride over and walk into the office with “Inbox Zero”. As a bonus, I get to see a little sunlight and feel like a surface-dwelling human! On days when I’m strap-hanging on BART, I’ll often just listen to podcasts. My tastes range from highbrow (We The People LlVE) to lowbrow (Pardon My Take), but it’s never anything related to tech or maps.
8:30 am
Having gotten settled with a coffee and a couple hard-boiled eggs from our office kitchen, I dial into one of our twice-weekly Telemetry team scrums. There are 7 of us on the team, typically spread around the world in SF, DC, Paris, Bangalore, and Singapore. This is a good opportunity to see everyone’s smiling face and check if there are any blockers on certain projects or to discuss upcoming issues to tackle. We don’t dive too deep on this call but it’s very common we identify conversations that need to be continued by 2 or 3 people offline later that day.
9:00 am
If I haven’t completely reviewed my digital to-do list on the commute in, I take a look now with perhaps some new context from the scrum. I use WorkFlowy for this because of its beautifully simple design. Then I settle into the first big block of “real work” for the day.
Lately, I’ve been focused on building out a pipeline that loads our speed estimates into an easily query-able data warehouse. Most of the programming for this project is done in Node.js modules that run in Docker containers on AWS’s Elastic Container Service (ECS), with heavy use of AWS S3 and Kinesis services. This all is made possible by the brilliant internal tooling of our Data Engineering team and will allow for much easier quality assurance and richer predictive analysis.
Other projects can be much more analytical, like discerning highway lane additions from GPS trace distributions. I entered the tech world with a PhD in Statistical Mechanics and very little software development skills, but in addition to statistical modeling work, I enjoy the more traditional “software development” tasks as well. As long as I’m learning something, I’m happy — and that happens almost every day at Mapbox 😀.
11:30 am
I start thinking about lunch. If it’s Monday or Wednesday, that means peeking at the #sf office Slack channel for the imminent “lunch bell” announcement. If not, then there are usually some pretty appealing leftovers from earlier in the week. Or if I’m striking out on my own I try to beat the Financial District crowds by picking something up around 11:45 am. I don’t really have a favorite spot; there are so many options nearby I feel like I could try something new every time. If I’m not too busy, I’ll sit down with whomever else is eating in the common area and have some social time.
12:30 pm
OK back to work. Maybe I jump back into my morning task for the rest of the day. It’s often around this time that one of my colleagues or I have discovered a complication in one of our plans from the morning. Maybe there is some technical limitation that we need to work around. Someone might start a conversation on Slack, as we like to communicate asynchronously when possible. It may well escalate to a live discussion for those in the same office, or even an impromptu conference call when discussing complicated issues with decentralized colleagues. Any design decisions or thoughts worth remembering are always added to the appropriate issues on Github.
1:30 pm
Now that we’ve gotten the “talking” out of the way it’s time to settle in for more coding. I try to stay focused by minimizing Slack and email windows for blocks of time. It’s safe to assume that if someone has an important item to communicate to you, they’ll DM you. But eliminating distractions is still an art that I have yet to master — plus a dose of cuteness from our #kids or #puppies channel never killed anyone, right?
2:30 pm
I grab some mixed nuts and, if I’m dragging a bit, a small coffee from our kitchen. We actually have plenty of more fun snacks than that (chocolate coconut chips, anyone?), but I’ve been trying to eat healthy for once in my life so I can chase Charlie around the soccer field for a good long while. The next couple hours are more of the same style of coding, with occasional design/debugging chats.
4:30 pm
Around now I start getting excited about seeing my family and start planning my escape route. On a fairly productive day I can leave now, taking a detour through the Embarcadero Y for a quick sweat on the elliptical. If things are a bit busier, I’ll probably just stay at the office until 5:30 pm or so before BART-ing home.
6:00 pm
My favorite time of the day — Charlie time! My wife Meghan also works so she and I have to fight each other for baby snuggles in the evening. Depending on the night I may play him some guitar, do some tummy time, or even give him a bath. His latest trick is rolling over — though he is still in “log mode” and has yet to fully master going in both directions.
7:30 pm
I wind Charlie down for the night with a bedtime story and some lullabies. His literary tastes aren’t too particular yet and we try to mix things up, but my personal favorite book is probably Hug Time. Any parent knows just how complicated bedtime can be. If the subject of infant sleep interests you there are about 800 books on the subject so I won’t bore you here with all our strategies. While I do that, Meghan starts on dinner — tonight it was a delicious BBQ chicken and quinoa thing she got from a food blog👌. We also get takeout more than we probably should.
8:30 pm
Dinner with Meghan — a great chance to catch up on our days, make weekend plans or deliberate on the next baby thing we need to buy (so many things!). After dinner we might watch a show together — depending on our mood, either a comedy like Veep or PBS NewsHour. One or two nights a week I’ll have some loose ends from work to tie up while she watches Fixer Upper.
9:30 pm
Nothing like a sleeping infant to get you into bed at a reasonable hour! We try to wind down around now. One of my favorite routines, right before my own bedtime, is giving Charlie his “dream feed” to fill up his tummy for as long as possible. Basically, we pick him up out of the crib while he’s still sleeping and stick the bottle in his mouth, while the milk just disappears and he barely wakes up at all!
10:00 pm
I like reading in bed but as an under-slept parent, I barely make it more than three pages. My current book is the genius Douglas Adams’ Mostly Harmless. It has a reputation for not his finest work, but I gotta say, I’ve been really enjoying it. After a modest dose of literature, I go to bed thinking happy thoughts about long baby slumbers. If I’m lucky he might make it until the morning!
Interested in joining Nick at work? Take a look at our career pages.
Nick on his new dad status and building accurate traffic predictions was originally published in Points of interest on Medium, where people are continuing the conversation by highlighting and responding to this story.