Categories
Confessions Musings Work

Four Years Later

It generally takes three times for the normal mind to understand something.
The first sign I got was seeing my senior coworker having part of his cube in
boxes, the second sign was having another coworker come by and shake my hand
saying, “it had been great to work together”, and the third sign was my new,
one on one meeting at 10:15. This is what it is liked to be laid off (although
I prefer the more humorous “budget actioned”). That was March 7, 2016 for me.
It was the end of a (just about) four year period of employment, my first
post-school job. I want to take the time, now a week or so later, to describe
what I actually did for four years. I want to give the highlights, the
lowlights, what I thought we did wrong and what I thought we did right. I will
preface this all by saying I harbor no ill will towards anyone I worked with or
Oracle itself. Oracle always treated me fairly, I am just sad it never worked
out for them.

May 28 2013 My first day of work. The week immediately
after Anime Boston. It was going to be the 27th, but that was a holiday
(Memorial Day). I was horribly late, the latest I ever was in my four years
there. I left early, but spent ages in bumper to bumper traffic. I called in to
let them know I was going to be late, but I was still majorly embarrassed.
After getting in late I was rapidly given a bunch of documents and a cube. As I
was getting setup in my cube, I remember opening the bottom drawer of my filing
cabinet and finding three liquor bottles in it. It was nice liquor too. I told
my coworker and he secreted it away never to be seen again. The rest of the day
I sat in a HR conference call. I would be the last legacy TEKELEC hire before
the company was completely Oraclized.

July 2013 My first real assignment. Before then I had been
familiarizing myself with the code, trying things out, going to meetings to
learn our process. Now I had my first real piece of work to do: S9. You can
read more about S9
here
. I was responsible for everything: GUI and backend. I would work on
this for the next few months. It would never ship as the customer who wanted it
never actually wanted it. Company wise, everyone was stoked that Oracle was
going to keep our office in Marlborough and not have us commute to
Burlington.

November 2013 I am a bit hazy about this point, but at some
point I finished S9 and moved to building the “Subscriber Activity Log”. A
feature that allowed a customer to track a subscriber as the subscriber’s calls
flowed through the system. You can read about this in the 11.5 release notes
as it actually shipped.

December 2013 My first company Christmas party. We go and
watch: The Hobbit: The Desolation of Smaug at the IMAX in Jordan’s furniture.
It was good, but my boss was mad it ended on a cliffhanger, so we went back
next year to see the final film. I remember spending a long time in traffic
trying to leave that area.

January 2014 Again, I am bit unsure when this happened, but
at some point I finished up “Subscriber Activity Log” and was moved to a new
project: MPG. The Mobile Policy Gateway (MPG) was a project to develop a device
that would deploy policies to a customers phone. One of the big use cases was a
provider could offload certain customers onto wifi if the load in an area was
too great. I would work on this for nine months, but it would never ship and it
would ultimately be cancelled. Company wise, things are going great. We are
making good money and selling a lot of product.

September 2014 NGCMP, our replacement for our current
configuration management platform (CMP) begins. I am project lead and I start
investigating new technologies. My first challenge is to pick between OJET and
LUX, two competing JS GUI libraries. I end up picking OJET, which
you can use today if you want. Weekly Wednesday calls with the offsite groups
(Bangalore and Nanjing) start. I went to JavaOne at the end of the month and
got this great photo:

I wrote more about that trip here.

November 2014 My first raise and bonus! I live the good
life. The NGCMP is starting to grow as other people finish up the projects they
are working on and get assigned to it. NGCMP starts driving major reforms
around how we do work. Code reviews are formalized, code quality tools like
SonarQube start to come into play, and we get serious about unit and robot
testing.

December 2014 Second company holiday party! We see the last
Hobbit film in IMAX. It is ok, but not worth sticking through the traffic in
that area.

February 2015 I miss most of the super bowl (I hear about
the Patriots winning during the flight) as I travel to India to meet the team
in Bangalore. I stay for a few days. I
wrote about this trip back then
.

March 2015-May 2016 Nothing of real note happens during
this period. I do get a raise, and we continued to make a lot of progress on
NGCMP. Customer demos are being planned at this point. NGCMP itself is in an ok
place. It has a lot of functionality, but a number of bugs. I produce the
greatest Star Trek TNG parody video during this time. It was a demo
highlight.

June 2016 Things are not going well for our business unit.
We have our first layoff. Management talks about pivoting to a cloud based
deployment. NGCMP gets “paused”. We no longer want to build a product that does
exactly what our legacy product did, we want to build something for a cloud
based customer. TREC is born and we pivot to new use cases.

October 14 2016 I close on my first house as more people get laid off
the same day. This would start a gradual local attrition of talent.

March 7 2016 Judgement day. I along with 90% of the office
get budget actioned. Layoffs are around 200 for my business unit. The Bangalore
team gets assigned to different projects and the Nanjing team gets reassigned
back to the legacy CMP product. TREC presumably dies as there is no one
assigned to work on it. The show ends.

That is the whole story. I really think things started to go wrong when
Oracle panicked and moved us to the “cloud”. It never made sense how we were
going to make money in that area and we certainly could not produce a viable
product for another year at least. In the end, I think Oracle could not take
the losses it was taking so it bailed. Ultimately it was a pretty good ride. I
had a lot of fun, made money, and learned a lot. It is just a shame it never
worked out.