An inside look at our FraudLab: 24 hours with Veljko Skarich

10/20 By IAS Team

Name: Veljko Skarich
Title: Principal Software Engineer, Data Science Fraud Lab
Joined IAS: June 2016
Degrees & certifications: B.A. Political Economy, UC Berkeley

9:00: I roll out of bed and read the news for 15 minutes, checking for any big activity in the security world. I usually skim the NYTimes, WSJ and the feed in my content aggregator, which is a collection of security and software sites.  After, I’ll often meditate or do some light yoga to clear my mind and face the day.

10:00 I plan out the work I want to get done for the day using a pencil and paper. I organize my to-do list by task size, and I try to do 2-3 big tasks and 7-10 little tasks each day.

10:10: I start checking on the status of tasks that I am waiting on from other people, and ping them if necessary for updates. If I need something from someone that day, I’ll kick off an email thread. I try to send my important emails early in the day so people can realistically get back to me that same day.

10:30: I  spend the first part of the day coding. I usually have a few feature branches I am working on in our fraud detection rules engine. The codebase is in Scala and is solely maintained by me. I work using a modified pomodoro method, where I spend 45 minutes focused on a task and then I take a 10 minute break.  

12:15: I work from home 75% of the time, so I will often do a HIIT workout before lunch. It usually lasts 30-40 minutes, depending on the routine. On the days I go into the office, I go into the gym. I like to workout earlier in the day because it gives me the energy to have a productive afternoon.

13:00: I eat lunch and attend our daily scrum meeting. I try to keep lunch light and protein based, to give me a steady energy release throughout the day. This is usually my first meeting of the day, since most of the team I work with is on west coast time, and I live in NYC.

14:00: Today I’m investigating an unusual user agent that is being flagged as fraudulent in our systems. I run some queries in our Hive cluster to figure out how our systems are classifying it. Usually in the afternoons I am in Slack conversations and I’m often guiding my code changes out to production.

16:00: I attend a meeting to go over the software architecture for a new component of our side channel analysis of browsers. Members of other teams are present, and we work together to make sure we have the same picture in our heads about how data will flow through the system.

19:00: I stop work for the day and usually get some dinner with friends. When I eat alone I tend to get takeout to minimize chores afterwards.

21:00: I work on some personal projects, such as a blockchain app I’m building, or a deep learning course I’m taking.   

01:00: I put away the screens and read books and magazines to unwind. Sometimes I’ll pour myself a glass of Rioja. The books are usually blockchain, political economy, or deep learning related. Some nights I don’t read, and instead work on creative writing projects.

02:00: Lights out and off to sleep.