Yes, I've always been in sort of math and science. Interested since high school. I went to college and started studying biology. Actually had some hopes of medical school somewhere in there on then I kind of lost interest in biology. Somewhere along the way, I started working for a lab when I was in junior and they had a lot of our in a data coming in. We're in a sequencing data. They didn't really have anyone realize it. So they set me up with Cluster in AWS and gave me access all the data. And from there I got my hands story with the plane. With the data learning the cloud environments and things like that, I found that was much more result oriented. So I like that a lot more than you know, Pipe adding all day long, um, loose college. It wasn't really looking for any jobs and biology and maybe some like bio analysis, gin up by the tech sector, Anything. But I started looking in the financial sector because that's where the majority of the data science positions were. Um, so I got a job with the consulting company. It's a small consulting company the CEO is actually a next colleague of my mother's, Um, so that worked out well in my favor, used my network there, but that was kind of the biggest experience that shaved my path, was getting in that lab when I was in junior and getting my hands on some data because I found that I just so much more enjoyed that basic lab biology.

uh um So typically so, I actually have sort of split responsibilities in my job. I work half a za data son into state analysts, and then the other half is business analysis component of it. So as a data scientist, you know the responsibilities range from data governance making sure the data gets ingested properly, making sure it's cleaned appropriately, making sure that none of your metrics air failing. Um, model maintenance is another thing, and they're making sure all your models are running cleanly. There's a lot of dashboards for that general infrastructure and environment up. Keep making sure the cloud environments air healthy and then also like doing some R and D. Just to make sure that all your processes are the most efficient. I think big part of the job is learning and evolving as you go. When I first started, we were on a really basic spark cluster for distributed on work, and from there we moved into the data breaks just because a lot of scales and doesn't cost as much money in the long run and now this as the result of a lot of research and development that we did internally and decided that was the way to go from the business. And analysts and the business. And analysts are mainly responsible for translating client needs into business requirements, which then get ticketed internally. So we said a lot of meetings with clients and make sure I'm fully understanding what they want out of their product that they're paying for. Um, I didn't go back and ticket. Take it using a ticketing software, those requests in as clear of fashion as I can to our developers, Um, and they build stuff out on the front and in the back. And, um, I want to test a little bit in there as well, for in terms of travel time, I used to travel along. We're obviously for all this pandemic started, but it was probably like, you know, once a month would travel. I'm generally within 100 or a couple 100 miles of where I work. Most of our businesses in the Northeast, you know, the Delaware Virginia main area, um, work from home. We had a remote infrastructure before the pandemic started because we're in the financial services. I'm this sort of data security that goes along with having a remembered infrastructure and also about had enough of our employees are already working from whom I used to go into the office. But when we moved to work from home for the pandemic, it was, you know, pretty seamless. We had a we have VM set up already. So he had to do is log in from your computer. Um, and it's generally the same hours. It's probably I'm cut back about an hour, so each day since we got into the pandemic, but it worked from home hasn't been bad.

So is the data scientists. The tools are, uh these are python on the basic statistical analyses software's. We try to shift more towards open source. One of our products extra revolves around moving people away from SAS to pipe down and are just as open source and distribute. It is I'm so much more worried, boss. So we focus a lot on play coming in all our when the world includes get too big, we need to distribute out into the clinic environment. The way we've had is using data bricks, which is sort of about really nice touchdown version of Spark um, which in which the clusters auto scale there's like native model maintenance. Integration with them help flow. You can also virgin your data using at a Delta tables. All of these air data Berks products. They're built into their sweet. So those are the big ones for data science. There's also exceed Well, obviously, if you ever need to query on older databases just to pull the data out in turn in order to model, you know, sequel as well, um, and selling power beyond It probably goes in there to power bi I and click um, in terrible at all the big dash boarding tools just cause you don't have to write custom scripts if you need a really basic analysis. Excuse me, um, for the business and this troll, there's zendesk injera. I would say the big too. So send us because the ticketing software that goes between the client in myself, you know, as a business analyst, That's where I basically make sure that all of the requirements are put into zendesk tickets on the decline it knows and understands and is aware of what he's getting. Um, and then whatever we push it deployment out to him, we can push those tickets back. For him to test on gear is between the business analysts and the developers. That's how you turn client facing tickets and the internal tickets to get the work done on them that gets test in the Q A department as well. Aside from that, that said, the big ones were like, you know, obviously Microsoft Sweet word Power point ghoul. Google slides any presentation software in there as well