Checksum Arcanius

Currently, I am sitting in the UW CSE building’s hardware lab, waiting for an email that will inform me that I once again have access to the side-room where the Robotics Capstone labs (which I am TAing) will be held. I was granted access before the quarter started, and my access magically disappeared on Monday, the first day of classes. So while I wait for the bureaucracy to catch up, I thought I would get caught up on my blog.

The holidays were a great time for me. Working as a TA meant that my work was totally done by Monday the 15th of December. When I was working for Microvision, my work load would actually pick up after school ended; this was the first time I’ve been at the UW where I actually had a work break during a school break. It was enormously fun. I got to hang out with roommates, my family (my brother was in town for two weeks), play in the snow, work on my own projects, cause trouble around town with all the friends that were in town, and generally have a great time.

This isn’t to say that I did no work between quarters — as I mentioned before, I signed up to be the TA for the Robotics Capstone course. I took the robotics capstone Spring quarter of last year and enjoyed it, but this capstone is significantly different — instead of autonomous slot cars like I worked on, this capstone is about distributed robotics systems. Instructor James McLurkin, a post-doc from MIT here at the UW, built over 100 small robots that together can execute distributed algorithms in the physical world. Some of their behaviors are quite spetacular — a physical bubble sort, clustering and dispersing, leader election and following, and so on. Watching a demo of the system is a fun and exciting experience. Getting back to the point, in order to be an effective TA for the class, I  “had the opportunity” have put in a lot of work over the break with the robots. Once you get to know your way around them, the robots are pretty easy to program with simple behaviors. The quarter should be fun.

At any rate, the rest of this post will have to come later, because I now have access to the room I previously mentioned.

My failure to convince Google to extend me an offer may have been the most fortuitous occurence that could have happened to me. Not getting a job offer from Google convinced me to cast my net further, and I have now decided to accept an offer from Facebook. The long and short of it is that starting in a little less than a year, I will be living in the San Francisco Bay area, eating warm lunches in the company cafe, and having a wonderful time learning and programming at one of the more interesting up-and-coming companies in the world.

I have a big decision coming up. It will likely be between choosing to start my work career at Facebook, where I’ve heard good things are coming down the pipe in my direction, and Amazon, where I have had an offer on the table since mid-October. The Amazon offer expires December 1, so that is more or less the timeframe in which I am working.

I feel pretty good about this decision because, even if I make the worse of the two decisions without knowing it, it will still end up being a pretty good decision. Both companies are well-respected places to start a career and both offer interesting engineering challenges that need to be addressed. Each job would also have its pluses and minuses, which I will have to consider carefully once i actually have the Facebook offer in front of me and I can make the most informed decision possible.

After what I would call a successful interview at Facebook, I met up with Scott and his roommate at a sushi place in Los Altos. I don’t have a lot of experience with Sushi, but I decided to give it a real shot and enjoyed it a great deal. Scott, meanwhile, was scheming up a last-minute trip to Boston for a friend’s birthday party. I ended up dropping him off at SFO before returning to his place to crash for the night.

The cold ended up waking me up this morning before my alarm, so I enjoyed a luxurious shower before heading north to Burlingame to meet with the Cloudera folks. What was originally slated to be a short morning turned into an all-day affair with me using their internet to finish up and turn in my team’s Hadoop assignment (which Vince went way overboard with, in a good way — examples will be forthcoming) between interviews with three of the four company founders. I had met previously with the fourth founder (or perhaps the first founder) in Seattle.

After leaving Cloudera, I travelled north to San Francisco in pretty ugly traffic, where I met with Martin. We caught up on recent life events and ended up eating at a quirky little restaurant called “Wierd Fish,” which is really a misnomer because the food was all excellent, well-priced, and really, not that weird. After food we considered a photography trip, but ended up instead meeting up with former classmates Sierra and Justin at a local bar.

At this point, I’ve been up since 6:40 am, so I think its a good time to consider sleep.

Tonight, Spencer drove me to Sea-Tac, I boarded a plane which took off, and then promptly returned to Sea-Tac due to a malfunctioning landing gear sensor. We changes planes, and eventually took off again about two hours late. I arrived in San Francisco after midnight, picked up a rental car and drove down to Palo Alto. Later today, I will be interviewing at Facebook. Tomorrow, I have an appointment with Cloudera. Since I am down here, I might as well check out Meebo as well.

This time, Google said no. :-(

This summer, I worked at Amazon.com as a Software Development Engineer Intern. When I started, I was very excited about the new possibilities and challenges.

Before long, however, my excitement gave way to what seemed like endless frustrations magnified by a bit of bad luck: my first month saw me miss two days to vacation, one and a half to being sick, my mentor away due to a newborn baby, the team member who knew the most about my project on a vacation for more than half the month, and a generally depleted team due to a sabbatical and two recent team member departures. One particularly memorable early team meeting consisted of my manager, one full-time employee who started in January, and two interns.

The result was that most of the time, I felt pretty on my own. Because some frustrations are normal with any Amazon new hire, my manager didn’t seem to think it was a big deal when I voiced concerns to him about my progress. Also, my manager initially thought the project would be smaller than it ended up; this led to more confusion between us, and on my part, a bit of distrust about my manager’s motives.

The first month was so bad, in fact, that there were times at work I actually considered just getting up and walking out on the whole thing, telling Amazon to forget it, never to come back, and just go on to enjoy the rest of my summer instead of dealing with the painful daily grind that my work had become. However, I had committed two three months, and I wanted the money too, so I decided to stick with it.

The worst case scenario, I thought, would be that Amazon wouldn’t want me back. That didn’t concern me too much, because at that point I certainly didn’t want to come back to Amazon; at least not with the team I had been working on. I suppose I did want Amazon to ask me back, but mostly just so I could revel in telling them “no” when the time came. Furthermore, there was the possibility of working for another group; all of my intern friends seemed to be having a much better time than me, so I mostly decided that it was just a group dynamics thing. I guess hadn’t totally given up on Amazon, and I didn’t want to burn any bridges.

Some time during the second month, I had made up my mind to stick it out. Occasionally, some breakthrough at work would actually get me excited to go to work the next day. Invariably, another problem came up in short order. Progress was excruciatingly slow. The entire second month went like this. A small bit of progress would lead to yet another problem. My motivation waned. I admit I didn’t spend 40 hours a day at work. Nevertheless, I did keep going back. Maybe my pursuit of the Bioengineering degree prepared me for this somehow: at times, both seemed fruitless, but in the end, I hoped, my time at Amazon would be worth it.

The second month was not really better than the first. While it did get me one month closer to the end date, it was the most frustrating month, because I kept hoping that at some point things would start getting better. When they didn’t I eventually evolved a sort of fatalistic attitude towards the whole thing: I would do my part, as much as my motivation made me capable, and not worry about how things turned out. This set me up for the third month of not really caring about anything.

Entering the third month, I was very much in countdown mode. My few successes were sullied by constant failures, both new and old. I knew my last day of work, and I looked forward to it with barely concealed enthusiasm. I continued to plug away at my project, and sometimes surprisingly, I continued to make progress. There was actually a point where I realized that if I stopped running into issues at such a prodigious rate, there was actually a chance (however remote) that my project might actually get finished. As it turns out, this didn’t happen (the coup de grace was another team’s public API that wasn’t implemented; I would end up writing and testing it for them), but nevertheless, by the end I was surprised by how much of the project was complete, given the constant issues I had encountered.

Indeed, as I approached my final week, I was thinking about the blog post that I would write about my experience. I had no particular ill will towards Amazon. I enjoyed everything about the company except the work itself — the interest email lists (particularly bikes-seattle@), the whitewater rafting trip, the Mariners game, the various lunches, the team socials, the Kart racing, the intern tech talks, the principals of Amazon talks — all of these were fun, interesting, and positive things. Around that time, Tim Montague posted about his summer experience at NASA. When I read it, I figured this post would read something like that. The “I learned a lot; I would not do it again,” sort of sentiment.

Then something amazing happened.

It started when my Manager’s Manager set up an appointment with me on Wednesday of my last week to talk about my Amazon experience. Around the same time, HR sent me a survey about my experience at Amazon. With that, the catharsis began. In filling out the survey just before my talk with my Manager², I tried to be as honest as possible. This really got my mind into the right reflective state for the meeting, and during that, I was able to get all of my concerns off my chest. It was an incredibly liberating experience, and it especially helped that the guy I was talking too actually cared about what I was saying, and wanted to know how the team could do better in the future.

Next, my manager set up a time to talk with me. The catharsis continued, and for the first time since somewhere in the first month, I felt like we were actually communicating healthily again. He apologized for underestimating my frustration, and I apologized for being complicit in letting the situation get so out of hand. We cleared up a lot of stale air during that talk; it was a deep breath of fresh air.

The two talks caused a lot of the bitterness that I had built up over the past three months to wash away. For the first time in over a month, I felt motivated to work hard. That afternoon, I completed an amazing amount of work, especially compared to how much I had completed over the last month. I cared again, and it felt remarkably good. I was disappointed that I wouldn’t be able to finishing my project, and even that normally negative emotion felt better than the indifference I had felt at noon.

That evening, I came home and wrote a one-line post that, as many of my one-liners do, confused a lot of people. Well, now you know.

The end result is that when my last day came on Friday the 12th, I was actually sad to leave. For a month I had been counting down the days; then for just over two days I wondered how it had gone so bad, and marveled at how an amazing Manager² can set so many things right so quickly. Today, I am happy to say can say that an offer from Amazon will compete with any other offer that comes my way.