1 [00:00.000 --＞ .260] Hey, three, two, one, there we go, we're live. 0:04
2 [00:02.260 --＞ .260] All right, so welcome Simon to Enterprise ML Ops interviews. 0:05
3 [00:09.760 --＞ .480] The goal of these interviews is to get people exposed 0:03
4 [00:13.480 --＞ .680] to real professionals who are doing work in ML Ops. 0:02
5 [00:17.680 --＞ .360] It's such a cutting edge field 0:01
6 [00:20.360 --＞ .760] that I think a lot of people are very curious about.
7 [00:22.760 --＞ .600] What is it?
8 [00:23.600 --＞ .960] You know, how do you do it? 0:02
9 [00:24.960 --＞ .760] And very honored to have Simon here. 0:01
10 [00:27.760 --＞ .200] And do you wanna introduce yourself 0:01
11 [00:29.200 --＞ .520] and maybe talk a little bit about your background?
12 [00:31.520 --＞ .360] Sure.
13 [00:32.360 --＞ .960] Yeah, thanks again for inviting me. 0:04
14 [00:34.960 --＞ .160] My name is Simon Stebelena or Simon. 0:01
15 [00:38.160 --＞ .440] I am originally from Austria, 0:02
16 [00:40.440 --＞ .120] but currently working in the Netherlands and Amsterdam 0:02
17 [00:43.120 --＞ .080] at Transaction Monitoring Netherlands. 0:01
18 [00:46.080 --＞ .780] Here I am the lead ML Ops engineer. 0:02
19 [00:49.840 --＞ .680] What are we doing at TML actually? 0:03
20 [00:51.680 --＞ .560] We are a data processing company actually. 0:03
21 [00:55.560 --＞ .320] We are owned by the five large banks of Netherlands. 0:02
22 [00:59.320 --＞ .080] And our purpose is kind of what the name says. 0:02
23 [01:02.080 --＞ .920] We are basically lifting specifically anti money laundering. 0:02
24 [01:05.920 --＞ .040] So anti money laundering models that run 0:02
25 [01:08.040 --＞ .440] on a personalized transactions of businesses 0:01
26 [01:11.440 --＞ .240] we get from these five banks 0:01
27 [01:13.240 --＞ .760] to detect unusual patterns on that transaction graph 0:03
28 [01:15.760 --＞ .000] that might indicate money laundering.
29 [01:19.000 --＞ .520] That's a natural what we do.
30 [01:20.520 --＞ .800] So as you can imagine, 0:02
31 [01:21.800 --＞ .160] we are really focused on building models 0:02
32 [01:24.160 --＞ .280] and obviously ML Ops is a big component there 0:01
33 [01:27.280 --＞ .920] because that is really the core of what you do. 0:02
34 [01:29.920 --＞ .680] You wanna do it efficiently and effectively as well. 0:01
35 [01:32.680 --＞ .760] In my role as lead ML Ops engineer, 0:01
36 [01:34.760 --＞ .880] I'm on the one hand the lead engineer 0:01
37 [01:36.880 --＞ .680] of the actual ML Ops platform team. 0:01
38 [01:38.680 --＞ .200] So this is actually a centralized team 0:01
39 [01:40.200 --＞ .680] that builds out lots of the infrastructure 0:04
40 [01:42.680 --＞ .320] that's needed to do modeling effectively and efficiently. 0:02
41 [01:47.320 --＞ .360] But also I am the craft lead 0:01
42 [01:50.360 --＞ .640] for the machine learning engineering craft. 0:02
43 [01:52.640 --＞ .120] These are actually in our case, the machine learning engineers, 0:02
44 [01:55.120 --＞ .360] the people working within the model development teams
45 [01:58.360 --＞ .360] and cross functional teams 0:01
46 [01:59.360 --＞ .680] ...