Not that my career is over, but I’ve been doing this long enough that I realized I’m starting to forget things I’ve done. I’ve had a pretty varied career, so I decided now is as good a time as any to record how I progressed from student jobs up to being a VP with a few asides for funny war stories that’s a little more interesting than my LinkedIn profile.
My first real job was a student job doing tech support for the residential internet at my college, NYU. College for me was the late 90s (NYU, CS major), so internet was just getting off the ground, wifi didn’t exist yet and only half the buildings were wired for ethernet and none of the student knew how to set anything up. Most fun part was that we did IP reservations per jack one at a time in a single BOOTP table that only one admin had rights to edit. We’d send him tickets and they’d pile up for weeks and we’d field angry calls about why they didn’t have internet yet. I heard they adopted DHCP right after I left. My best lesson from that job was working with this one student who didn’t understand a thing I said. Everything I told her to do she answered with “What?” Over and over. After talking to her I made sure to make fun of her to all the other techs. We made fun of everyone who called in so this was nothing new. When I checked her ticket status a few days later another tech had left a note, “She is hard of hearing so please speak loudly.” I stopped assuming everyone else was dumber than me.
My next gig during a summer break was actually one of my most interesting. I worked for a professor at Rockefeller University on a project funded by a grant from the Human Genome Project. We had this data set of genetic markers harvested from a huge family of Mennonites. The family had lots of individuals across multiple generations alive and willing to give samples for sequencing so we had a big enough data set of related people to see heritable markers chopped up dozens of ways. The basic idea was that because genes migrate in chunks, the more often samples showed overlapping sets of genetic markers, the more likely those markers were near each other on a strand. Using some fairly complicated math you could make some strong statistical inferences about the complete sequence. We publisehd the data, modeled the formulas in Java and hosted a website where geneticists could test some hypotheses and get estimates on likelihood of sequences being valid. It was interesting and complicated work that hopefully helped advance some science and it earned me an author credit on scientific paper. That was a thrill and something I was really proud of.
I had another part-time job while in school doing HTML for a book publisher, Basic Books. Not terribly interesting, but it was in that weird moment in time where a college senior who owned a copy of Webmaster in a Nutshell was basically the entier digital team for a decent size company. I was just freehand coding HTML and designing completely ad hoc. I remember a book editor showing me an animated gif he liked that he thought I should use and the CEO telling me to make the logo “more electric” and not having any idea what he meant.
My first real, real full-time job was a consulting agency called Quidnunc. We worked for a bunch of clients who were trying to get their first website online or pitching some ridiculous startup idea. Nobody knew what they were doing and it was great. I also had what felt like an unbelievable amount of money for a 22 year old and the best part was being offered paid relocation to the UK for reasons I never quite understood. But I got to spend over a year there working on some fun projects. We built the first DTC car-selling site in the UK which is now sadly defunct, but it was incredibly ambitious at the time. We did some projects for a liquor retailer, Cambridge University, a utility company. A lot of Java, but also Perl, VB and some other stuff. There was a lot of frothy excitement in those days and I got to work with a lot of smart, but very young and inexperienced people. I was made a manager at around age 24 and didn’t do very well. We drank a sickening amount of lager after work every day. After 18 months of beer and no sunshine, I was ready to come home. The company was melting down in the web 1.0 bubble burst so I took a voluntary layoff and moved back home.
My next move is the one I most regret. I had a bit of real job experience and my CS degree was fresh in my head so my father suggested getting a masters and I should have done it, but instead I went on a few interviews and got an offer relatively quickly. One of my consulting gigs at Quidnunc was using a crappy enterprise CMS product called TeamSite. It was one of those products that nobody liked, but it managed to sell itself to a lot of big enterprises. I ended up taking a gig at HSBC and was there for over two years. TeamSite was primarily managed via Perl which was still viable at the time. We had like a 10-person web team and very little direction. Half the team had no coding experience, the other were mainframe people inherited from an acquisition. They were all nice enough, but I was languishing. I received the most demoralizing year-end review of my life when my manager (a great guy, but a non-technical PM) told me that I my tech skills were “complete” and could not get any better. It was high praise but also completely dispiriting because I knew it wasn’t true and I’d definitely hit my ceiling at this org. Work was boring and was mired in becoming a product expert in a product I didn’t like. My favorite anecdote from working here was that it took two days to get a PC provisioned and in the days before laptops were common it meant that I spent two whole days at the office sitting at an empty desk. All I had was a phone which I used to talk to my girlfriend like three times a day. I was interviewing on and off less than 6 months after starting, but I needed to wait for the recession to thaw to actually get a better offer. I married my girlfriend and quit the job in the same month.
I landed someplace a lot more palatable, the City of New York Dept of IT. Unfortunately, it was still an agency heavily invested in TeamSite, but that’s what got me hired and broke me out of my slump. Work was a lot more engaging and interesting. We ran over 100 agency websites and the 311 call center out of a single CMS with a small team and talented people. The office had a ton of investment from Mayor Bloomberg which was split between FTEs and an absolute mob of consultants from Accenture. Our team was blended. Accenture was pretty crappy partner. They used some underhanded techniques to make themselves indispensible. They also engineered a release process that took almost 8 hours so it had to be run overnight once a quarter. It was torture. But I really liked it there and felt good working for a public service. During the transit strike we had to be on site monitoring systems round the clock in shifts. I walked to office across the Brooklyn Bridge at 11pm for a graveyard shift and took a nap on the floor. Kinda gross, but I was still young and felt like I wasn doing something valuable. But eventually the relatively low pay and being stuck in the world of TeamSite for nearly 5 years was wearing thin so I started looking around.
I ended up at Razorfish (now rolled up into Publicis.Sapient) with a hefty raise. Razorfish was a big digital agency that had sorta survived the bubble burst and still exists in the form Publicis.Sapient today. At the time, the market was heating up again, there was a lot of investment and a lot of meaty projects to work on. I was at this point extremely efficient at Perl and TeamSite, but my generalist skills were not sharp. Luckily, Razorfish had a huge TeamSite client (Conde Nast) who needed a lot of work. I spent almost two years working on sites at Conde and it was pretty great. I met one of the founders of reddit very briefly while he was walking around the office. As much as I was tired of TeamSite, it got me hired. And being an agency, we would have to jump from client to client and different tech stacks frequently. I was back to doing mostly Java work with some later clients. Ford, Mercedes, EMC, PNC, Dow Jones. It was around this time that I got my first real experience with tech leadership and agile process. I got accustomed to Scrum doctrine and saw it’s value. I also witnessed an agile coach show up and grind our progress to a halt. It was terrific experience in seeing how the good parts and bad parts are so superfically similar and what elements of agile are actually important. The first time I actually had to be the scrum master, I remember the program manager just telling me “Here’s your team, good luck” and just being thrown in the deep end. It actually went great and by midway through the project we had a sprint burndown chart that was a perfect 45° line. It was also my first time working with an offshore team as most of the devs were in Argentina. It was a bit of a shock at first, but we hit a rhythm pretty quickly and really relied on IM and using a digital backlog system instead of post-its. That was a big shift at the time as much it’s unthinkable to do it any other way nowadays. I was at Razorfish for nearly 7 years altogether during which I had two kids, took a sabbatical, and weathered another recession. It was mostly a great time and probably saw my biggest career and personal development but eventually I decided to move on. Interestingly enough, a lot of my peers are still there some with tenures beyond 20 years at this point.
I really wanted to escape agency at this point because travel and client demands were grinding me down while I had young kids, but somehow I got pulled back in. A lot of ex-Razorfish people ended up at a competitor called Huge and I was invted to apply/poached by some former colleagues. Indeed they had a short-term need for TeamSite work which was a bit disappointing, but ultimately it worked out very well. I filled a lead role and got to hire people into execution roles. In it’s prime, Huge was absolutely top notch operation. They had a razor sharp execution process and top quality talent all around, especially design and UX. This is where I learned what great QA process looks like from some incredibly dedicated testers who let nothing slip and knew what made a good product. I got to work on some really impressive projects for Lexus, Morgan Stanley, Toys R Us (RIP) and see frontend technology really take off with some experts who knew how to use it. A lot of clients who previously would have opted for TeamSite were now looking at AEM. I had a lot of success building with it, but was a bit worried about falling into the product specialist hole again. But it did take me directly to my next career twist which was to transfer to a sisten agency at my manager’s request.
Elephant was spun up down the hall from Huge. It started life as a conflict shop working on clients that were direct competitors to Huge’s clients. The first dozen employees in Brooklyn were all from Huge and the whale we’d landed was Goldman Sachs. They had bet on AEM and liked what we’d done for Morgan Stanley. I spent a little over a year working with GS and some other partner agencies getting their first consumer-facing product off the ground. The product was an all-digital consumer lending platform and the brand was going to be marcus.com. I have never before or since endured so much existential dread of getting up in the morning, but somehow powered through the whole thing. At this point, Huge and Elephant were sharing dev resources in Colombia but had separate teams in the US. We had dozens of people working on this account with a ton of uncertainty and maximum pressure. Both for us as a new venture and the client who were incredibly intense and demanding. They were trying to do so much at once, it was a miracle we pulled it off. They’ve booked billions in loans since the launch, so it’s my most economically impactful project of my career by far. It certainly thickened my skin quite a bit too. There some enjoyable moments amongst the misery and the team made some strong bonds. Once that finally wound down, I pivoted to some other clients which wore me down in different ways. I started to split my time between Shaw (Canadian telco) and Comcast. The problem with both was travel. Comcast was really interesting work. We were doing in-store digital experiences for flagship stores. This my first foray into this space and it was fascinating. We partnered with fabricator and AV techs to build experiences with screens, projectors, lighting controllers and custom-built devices meant to be immersive and exciting. The constraints of this kind of work were completely different from building websites or apps with the biggest constraing being the physical space. We had to have space to build and test prototypes, but also there was the load in and installation which took many weeks and had to be done on site at stores in Washington DC and Pennsylvania. Done concurrently to giving facetime to Shaw in Calgary, CA. A real low point at this time was flying twelve hours to Calgary for a meeting, going up to their (really beautiful) office when the stakeholders were at the back office on the other side of Calgary. We did the whole thing over the phone from 10 miles away instaead of 2000 miles away. Meeting went great though (and I talk about this everytime we debate remote vs in-person work policy). The Comcast projects ended up going very well. I was a director at this point and doing a ton of client relation work, contract negotiation and project planning but also keeping my nose in technical work. Unlike a pure digital project, we were at the mercy of actual physical construction. We did our first round of field testing and setup with hardhats on because they were cutting ducts and installing the floors while we worked. First day I walked on site I had a crew of construction workers asking me questions about installing a projection screen and I had the answers. Younger me would have shrunk away from answering something that wasn’t my job, but I at that point I knew that I knew as much as anyone else so let me help these guys out. It will help me later on and it did. We got along with all the work crews really well. I had to bark at the AV installers for lying about progress many times. The final product was incredible, but I was once again exhausted and missed my kids.
West Elm/Williams Sonoma taught me one really valuable lesson: It’s better to be too busy than bored. At West Elm I was booooored. I wanted to slow my pace of work and overshot by a lot. I spent a year in their absolutely immaculately decorated office looking for stuff to do. It was horribly awkward and I never had clear expectations of anything set for me despite asking point blank several times. All of the people who had impressed me enough to join up ended up departing before my year was up, so I took the cue and left also. I liked the folks and I got at least two projects shipped which isn’t bad, so not a total loss. I also developed a lifelong aversion to any job title with the word “Architect” in it. On paper, it was a dream to get paid so much to do almost nothing, but it gave me absolutely gnawing anxiety.
So I plied my network one more time and it was a lot simpler this time. I texted my old boss and she had something for me right away so ended up back at Huge once again. They had another big client doing digital retail only this time it was in Asia (SK-II, division of P&G). Within a month I was in Singapore (longest direct flight in the world at the time, but it was business class) helping setup a popup store built from the ground up on Orchard St. This was a bit scary because I joined late in the project and had nightmares from the last time I’d done a similar project like this, but it actually went really smoothly. We fixed a load of on-site issues, but the construction and installation was all like clockwork. Singapore is an underrated tourist destination although I highly recommened you visit with a corporate card to cover all your expenses. I ended up doing a load of crazy projects for this client and spent a lot of time in in the air. While we did a lot of experimental experiences with eye tracking, scent dispersal, AR, gesture tracking, RFID, transparent displays and robots and they just kept paying us to come up with stuff. My job now included a load of contract writing, vendor management and even hardware procurement. At some point the money facuet slowed to a trickel and I went back to the world of CMS and web. Going back to building websites (Northern Trust, United Air) felt like a letdown but it was also more stable and predictable. Like when I predicted our contract terms were grossly insufficient to pay for all the work they needed and that were dependent on prerequisites from a client who couldn’t answer an email in less than a month. I very accurately and clearly assessed how hopeless it was every chance I had, but we still ended up blowing our budget by a mile. Then the pandemic hit and the client could barely pay bills let alone approve an extension. I got laid off along with most of the tech staff at Huge around the time we cut off work with that client. It was my biggest failure even though I knew I did everything I could. The site ended up launching over a year later.
This seemed like a decent time to chill out. There was a ton of people on the market and the pandemic was out of control. Kids in remote school were home all the time, my wife was still working full-time so no pressure on me. I had some interviews but didn’t get far. My network came through again with some tasty bits of freelance work. I helped an old colleage get his agency, Mojave off the ground. It was fun work and low stress.
Just when I started to relax a bit a real dream job showed up on LinkedIn and I applied. I’ve always had a keen interest in news and was hoping to work someplace that was more fulfilling that all the corporate clients I’d dealth with in the past. NY Public Radio needed a new head of tech and it was perfect fit. My team has a lot of autonomy and the right amount of scale to do interesting things for a sizable audience but without a ton of bureaucracy. I directly manage managers but I still have personal relationships with everyone in my org and can also get 1:1 time with anyone in the C-suite when I need to.