Wednesday, 17 May 2017

Podcast Interview with Darko Križić, Prodyna

Another stupidly late podcast publication on my behalf. Somewhere early March (yes, I KNOW - dammit!!!) I had a great conversation with one of our prime Neo4j partners in Germany and across Europe these days, called Prodyna. We did a couple of events together, and I found that some of their thinking and case studies really aligned very well with my own. So we got together for a chat. It's a bit annoying because both of us were referring and looking forward to GraphConnect - and I clearly missed that deadline/timeline. But still wanted to share the conversation... Here it is:
 As per usual, here's the transcript of our conversation:
RVB: 00:02.689 Hello, everyone. My name is Rik, Rik van Bruggen from Neo Technology, and here we are again recording another podcast, a little bit closer to home. It's actually a really special podcast for me because it's exactly two years ago since we started it on request or instigation of my dear friend Michael Hunger. And this week we've invited someone from Germany in order to talk a little bit about of the wonderful things that PRODYNA is doing with Neo4J. And that's Darko Krizic from PRODYNA. Hey Darko, how are you? 

DK: 00:40.222 Pretty fine, and excellent pronunciation [laughter]. It's great. 
RVB: 00:44.628 I make an effort. I try to make an effort to pronounce everyone's name correctly. It doesn't always work. But thank you for coming online Darko, it's great to have you here. And well, I've been watching you guys for quite some time and we've been developing a partnership between Neo4J and PRODYNA, quite aggressively recently. But most people won't know you guys at all, so why don't you introduce yourself? 
DK: 01:08.868 Yeah, so my name is Darko Krizic and I'm co-founder and CTO at PRODYNA. And PRODYNA originates from Germany and now we are active all over Europe. And we have about 280 employees. And yeah, we are doing more and more in the fantastic world of graphs and especially Neo4J. So we have some active customers and we have a lot of developers who are really doing a lot of very interesting projects with Neo4J. 
RVB: 01:43.896 Very cool. And PRODYNA, I mean is it fair to say it's like a high-end, very specialized system integrator? That's kind of what you guys are, right? 
DK: 01:52.106 Yes. Actually software development, software architecture, integration, consulting, project management, UX design, and all around this topic. 
RVB: 02:06.384 I've seen some really great cases why you guys for German and other customers, and that might be a really nice segue into asking you my favorite question really and that's why did you guys get into the world of graphs? What's behind it? What attracted you to it? 
DK: 02:26.129 Yeah, I like this question. I mean, at PRODYNA we have a business unit called Data Science, and data science is about managing big data, complex data. And if you have a normal customer 90% of our job is to create applications that take data from end users or from other systems, store this data, and quickly finds this data. So what we are doing is we are creating database-driven applications mostly. And we are always monitoring the markets for interesting technologies that can ease our work, and we have knowledge for relational databases which are on the market for 40 years, but also very interesting NoSQL databases. And in the last years, we found more and more that there are graph databases and so we started to investigate and found out what are the advantages of graph databases. And, yeah, this leads to the fact that we are using-- we selected Neo4j as technology. We are partners of Neo4j, and we are doing more and more projects with Neo4j. Even for new projects instead of relational databases, we instantly start using Neo4j because it is much, much easier. And one day I was out on the WJAX in Munich and there was the booth of Neo4j, and I met personally Stefan Armbruster. I was talking with him. I invited him to an internal session at PRODYNA and asked him if he would like to introduce the technology to a wider audience in PRODYNA, and that's what he did, and people were really amazed. So we have a lot of people who are really skilled in Neo4j and are really eager and bringing the wisdom to the rest of the company.
RVB: 04:32.529 Very cool. Was there a particular use case that you thought was a great fit for Neo4j, or a particular problem that you thought would be easier solved with Neo4j? 
DK: 04:42.460 Well, as I said we have a lot of customers, a lot of database requirements. We always have a domain model that might be really complex. And in the last years we saw that those requirements of the customers are getting more and more complicated. Customers want to be agile. They quickly come up with new requirements and in traditional databases it might happen that you have to restructure your complete database to fulfil the new requirements. And therefore, short answer, yes, we have requirements or we have demand for an easier database technology in order to easier and quicker fulfil the requirements of our customers. 
RVB: 05:29.206 Oh, very cool. And I think I saw at one point, I saw you guys present about one of your customer cases which was in biotech if I don't mistake myself here. 
DK: 05:41.241 Yeah. Yeah. The customer is called Adama, it's a worldwide active company. And they have an interesting use case where they actually are-- they are selling products for crop protection, fungicides and herbicides and they bought about 50 local companies in the last years. And they lost overview of what the product they actually have and which product are registered for which kind of treatment in which country. And they found us as a partner of Neo4j. So we analysed their data or their multiple data sources and we created a great application for the customer where all data and this knowledge of the company is stored in a single Neo4j database as a single graph, and which allows the customer to do a really great insights. It gives the customers greats insights on products and registrations and compare countries, compare products and so on. 
RVB: 06:49.048 That sounds great. I think we might see that presentation at GraphConnect, isn't that case? I thought it was on the agenda, I'm not sure.
DK: 06:56.219 Yeah, yeah. For my knowledge this might be-- I'm not sure if this is going to be presented or another use case you're doing is fraud detection combined with machine learning. So-- 
RVB: 07:06.052 That's cool, very cool. 
DK: 07:07.240 Yeah, this is another interesting topic we are working on. But yeah. I need to clarify that actually [laughter]. But, I mean we [crosstalk]-- 
RVB: 07:15.674 Sorry. I jumped the gun a little bit there. Sorry. 
DK: 07:19.109 No problem. I mean if we are at GraphConnect and if a customer wants to talk about this use case we are there and we will talk about this, independent if we have a great presentation, so I don't know. But yeah, we are there and feel free to contact us. 
RVB: 07:35.999 Absolutely. And Darko, maybe, I always ask this question to people that are in our industry and are actively developing and moving the industry forward, people like yourself. What does the future hold for you guys? Where do you think this is going? I mean obviously, we see a lot more adoption happening across lots of different industries, but are there specific things that you think will really impact either the adoption by customers, or the ease of use, or the product, or whatever? What does your crystal ball say? 
DK: 08:11.759 I personally have the feeling that about 40 years we got relational databases. Before that, we had some things like non-relational or hierarchical databases. And in between, we had very interesting concepts of wide column databases and document stores like MongoDB. But they bring some advantages but a lot of disadvantages because, for example, you need to think about how I'm going to query my data later. And this is against the idea of being agile. And for my personal opinion, graph databases are now-- they bring superior modeling concepts. So any kind of information I have like relational data, or hierarchical data, or topologies, ontologies, whatever, it's very easy to map this into the database. And so I personally think that graph databases will become much, much, more important in the future and I don't know if they will replace relational so quickly but I can imagine that maybe 5 or 10 years ahead about 50% of all databases will be graph databases. 
RVB: 09:20.851 Oh man, I think, of course, you're preaching to the choir here, but I can't wait for that to happen as well. So that's a really, really, great perspective. Are there any specific things in the product that are your favourite feature or something like that in Neo4j, or what's the thing that you like most? 
DK: 09:42.428 I like Cypher very much. It's so intuitive and easy to learn, and you can really show somebody queries in real time. And what I also like very much, and this is very important to Neo4j, it has a great community. So it's open source, so we also did some pool requests or some changes. For example, we would like to be that the mapping part, OGM, is also running in an OSGi container which is important for some of our customers. And this really works and there are a lot of consultants out there who know a lot of Neo4j and this makes this database technology extremely powerful and a bright future. 
RVB: 10:28.225 Fantastic. Well, thank you so much for sharing your perspective there, Darko. I really do appreciate it. I think we're both onto something here and I look forward to working with you guys and seeing you guys at GraphConnect to move this entire industry and Neo4j forward I guess. Thank you so much, Darko. It was a great pleasure talking to you. 
DK: 10:50.729 You're welcome, it was also a pleasure for me and see you on the GraphConnect, right? 
RVB: 10:55.887 Fantastic. Thank you, Darko. 
DK: 10:58.170 Thank you, Rik. 
RVB: 10:59.133 Bye. 
DK: 10:59.272 Bye.
Subscribing to the podcast is easy: just add the rss feed or add us in iTunes! Hope you'll enjoy it!

All the best

Rik

No comments:

Post a comment