OpsGuru Meeting Referred by AWS

Source: Notion | Last edited: 2023-10-31 | ID: 1f1399c4-159...

Hi Terry,

Great connecting with you yesterday.

As discussed on the call, sharing the details of Engineering on Demand and Customer Reliability Engineering which we went over together. Also including our One Pagers on the programs as well.

Engineering on Demand:

60 hours of hands-on keyboard work from OpsGuru Engineers
Architecture Reviews, DevOps, Cost Ops, Service Adoption, etc.
Out of pocket cost to EonLabs is roughly $2,500 CAD Customer Reliability Engineering:
24×7×365 on-call supporting engineering response for your critical services running on the public cloud
OpsGuru will need to dive into EonLabs’s needs to provide a cost estimate Please let us know which program would be a better fit for you once you connect with your team, and then we can discuss next steps.Talk soon,

Video

PDF

augment://mtg?tid=teBZFmjtnixW

Key Takeaways

• The company heavily relies on AWS and open source machine learning algorithms, outsourcing some jobs and using AWS managed services. They plan to hire more people for better separation of roles.

• OpsGuru is an AWS Premier Tier service partner with three main technical focus practices. They offer Engineering on Demand and partner with ISV partners to fill gaps in customer needs, focusing on understanding each individual customer’s use case.

• The organization has experience working with financial institutions, fintech, and cryptocurrencies. They offer customer reliability engineering (CRE) for 24/7 uptime monitoring and incident response, and engineering on demand (EOD) for proactive DevOps and infrastructure improvements.

Current Tech Stack Team

The company is heavily dependent on AWS and open source machine learning algorithms. They outsource some jobs and use AWS managed services. The in-house team managing AWS consists of a single person, and they plan to hire more people for better separation of roles.

OpsGuru** Overview Offerings**

OpsGuru is an AWS Premier Tier service partner with three main technical focus practices. They offer Engineering on Demand and partner with ISV partners to fill gaps in customer needs. They do not have physical SaaS solutions but focus on understanding each individual customer’s use case and building what they are looking for.

CRE EOD Services

The organization offers customer reliability engineering (CRE) for 24/7 uptime monitoring and incident response, and engineering on demand (EOD) for proactive DevOps and infrastructure improvements. EOD is offered at a special rate of CAD 2500 for 60 hours for the first three months. The longer the contract and the more hours, the lower the hourly cost. A follow-up email with more information will be sent

File attachment

10:08:50 AM - Speaker A

Quickly. If you could, if you could just give us a quick overview of what you built to date in terms of either within AWS or even on Prem or in other cloud providers.

10:08:59 AM - Speaker B

We can heavily relying on AWS because we always had a kind of bias or mandates that we don’t want to invest in something that doesn’t although they may sound cheap in the long run, but technology is changing so quickly we rather use the latest and just to wow, that sounds about right. When we are working with AWS, as technology evolves, we don’t have to keep up with all the latest administrative, technical administrative stuff. So it’s been doing well for us and even for some of the heavier lifting of machine learning training we were putting AWS computational powers to use as well. So that’s the main focus. In terms of the machine learning algorithm that we use, we are basically using open source stuff from TensorFlow to models such as LSTM to the latest Transformers. We are using these kind of things. So yeah, nothing too fancy or propagatory. Yeah, just the open source stuff got you.

10:10:22 AM - Speaker A

And it sounds like you have a relatively lean team internally, which is fantastic. Just out of curiosity, are the contractors you’re outsourcing to today specific to DevOps or infrastructure or what is like, I guess who is?

10:10:35 AM - Speaker B

Yeah, good question. So in terms of outsourcing development okay, so to be very specific, we have one specific oh, we currently have two specific jobs that are being outsourced. Both of them are kind of similar. One is Simulated Exchange and we found another outsourcing partner that’s in open source space, but we signed an NDA for now. We haven’t been able to disclose them, but we found them on GitHub and then we contacted them and they say, hey, they can do commercial work as well. So we contract out some of our infrastructure stuff to them. To them because although we are strong on machine learning, it’s just that we are relatively a small team and we are trying to hire more people. And when we were trying to hire more people, we found that, hey, somehow our current structure is too tightly knitted. We cannot do the separation of rows too nicely. So we were contracting out to some of these experienced people who were experienced in building financial infrastructure to help us to build a more robust model so that we can have a better separation of roles, so that we can hire more people to have a better I’m not sure.

10:12:06 AM - Speaker A

No worries. And I guess obviously the questions are coming from more around our engineering on demand offer specifically. So trying to frame up the thought process of my head here, which is great. Andrew, did you have any other questions around the tech stack specifically before I jump into more of a high level Oscar overview and obviously business discussion?

10:12:26 AM - Speaker C

Yeah, certainly. So from the technical standpoint it sounds like you’re standardized on AWS right now. The tooling is the open source components. Are you using the managed services through AWS? Are you using the Sage maker and that sort of thing? Or is this kind of infrastructure that you guys have deployed that you’re managing to run those models on, like, with attached GPUs or what does that look like? Do you have any sort of background on that kind of footprint right now?

10:12:58 AM - Speaker B

No, mainly the server side. Well, at most, like, I guess we try to use what’s that called? Is it serverless? No, with lambda. Yeah, with lambda.

10:13:14 AM - Speaker A

Yes.

10:13:14 AM - Speaker C

Okay, great. And the support that you have around the infrastructure and the AWS account side, is that a team that’s in house, or is that part of that? Is that open source? I know you talked about? Oh, sorry. I meant contracted.

10:13:34 AM - Speaker B

Yeah, good question. So, for AWS, we all in house things that contracted out is non related to AWS. It’s just that they are building the stuff based on rust and to increase the performance. And they have a better structural hierarchy so that they build it first, and then we will have to put them to our AWS ourselves later on.

10:14:00 AM - Speaker C

Okay, great. So the team in house that’s managing the AWS, is that a few engineers? Is that a single person?

10:14:08 AM - Speaker B

Basically? A single person.

10:14:12 AM - Speaker C

Okay, great. Yeah, I think that’s excellent. When it comes to the ML space, there’s a ton of obviously deep discovery that can be done to fully understand, but I think this paints a picture for now. So, Scott, I’m happy with what we have here. You’re on mute.

10:14:33 AM - Speaker A

Yeah. Thank you. Awesome. Thanks, Edge. And so, Terry, did you want me to give you a quick overview of offscrew as an organization just to set the stage for the conversation and also give you a high level overview of the engineering on demand offer we have?

10:14:45 AM - Speaker B

Oh, yeah. Especially, what kind of client have you been mostly working with and those sort of stuff. Yes.

10:14:52 AM - Speaker A

Yeah. Fantastic. Let’s dive right in here. All right, so can you go see my screen here?

10:15:00 AM - Speaker B

Yes.

10:15:01 AM - Speaker A

Perfect. Great start. So Terry Oscar is an AWS Premier Tier service partner within the AWS ecosystem. That is the top tier that they actually give out. Last week at Toronto Summit, we were named the AWS Canadian Partner of the Year, which we’re very proud of. For the past three years. We’ve won Consulting Partner of the Year and then Migration Modernization Partner of the Year. And now we’re actually the Overarching Partner of the Year, which we’re very, very proud of. So putting these on the screen is really not us bragging about the award, specifically award. And this award, it’s just a showcase to customers who have never heard the name obscure before. That our delivery team is absolutely phenomenal at what they do, and the expertise they bring to the table is almost unmatched. Right. And so quick background. Obviously, I’m not going to bore with the history lesson here, but we’re founded in 2017, we’re acquired by a company called Carbon 60 in 2021. Both are Canadian based firms as well. Obscure was founded in Vancouver. We obviously have a large presence here in Canada, some folks south of the border, and then two large development hubs overseas in Israel and Romania. And so three main technical focus practices would be so when I was at AWS, had partners come to me and say, hey Scott, we can do basically anything for anyone. And I always got a little bit weary of that statement there because I’m a firm believer that not everyone can be an expert in everything. But although this is not everything that Opscrew can do for our customers, these, I would say, are about 90% of our revenue comes from one of these three pillars on the screen here. Right. So the first is infrastructure and DevOps. I mean, relatively straightforward here, but building out your foundations on AWS, doing any type of infrastructure, architectural design work. We do a lot of migration work from other cloud providers into AWS. Security and governance is obviously a huge focus of ours as well. Just being in the cybersecurity space, cloud native development would be the second focus practice as well. So that’d be things like microservice transformation containers, kubernetes serverless computing, elastic computing, that sort of thing. And then the third one, which may or may not be interesting to you, is really around the data and AI space. Right? So we’re building out some of the most complex data lakes in Canada to date with our enterprise segment nonetheless. But we do a lot of AI ML work. We always tell our customers that we’re not the business logic of ML, we are everything in the background to kind of get the AWS services up and running so you can help essentially build your algorithms and everything else you do on top of that. Right. Another thing that we not many customers are familiar with to date, but it’s actually one of Amazon’s competing tools to Tableau. It’s called Amazon QuickSite is the data visualization tool. And so it’s incredibly powerful in my opinion. Obviously it’s just an opinion, but more powerful for less money overall than Tableau. And so fantastic tool overall, but obviously that’s up for interpretation as well. So, three main pillars engineering on Demand would really focus the majority of the efforts into this infrastructure and DevOps pillar specifically. But if there’s anything in any of the other pillars that you wanted to explore with us, happy to dive in. And really, again, the goal of this was just to kind of set the stage of who we are, a little bit of background, and then hopefully dive right into the Engineering on Demand use case. Right. Any questions on any of this stuff, Terry, before we flip gears into Engineering on Demand specifically?

10:18:12 AM - Speaker B

Yeah, perhaps it’ll be nice if you can share with us some of the because throughout the period of you servicing some of the clients, perhaps there’s something that you can productize or made it into a more full blown product that you can make available for future client to use. So if there’s something like that, we would like to take a look as well to see if there’s something that can fit us so we can pay and don’t have to contract it out or do it ourselves.

10:18:48 AM - Speaker A

Yes. So in terms of productization, we are almost strictly a professional services company. We have partnered with a few ISV partners as well that obviously we look at this overarching customer needs and then we’ll build essentially our ISV playbook from there to help fill some of the gaps that we’ve seen in the years of conversations we’ve had with customers. So we’re very strategic with which ISVs we partner up with. But in terms of actual products ourselves, our implotations products are more so service based, so engineering on demand would be one of our products or offerings, whereas we don’t have a physical SaaS solution, for example. Right. That’s just not the business we’re in. We go in, understand every specific customer’s use case and then really help them build what it is that they’re looking for. Obviously taking into account the cost optimization and reliability and scalability and everything else that AWS has to offer. But we really start it’s a traditional AWS work backwards methodology, but understanding each individual customer. And I think the key to that, Terry, is really at the end of the day, you as the customer own the IP to whatever it is that we build for you. It’s really just you’re contracting out the human capital piece of this to Ops guru you’re left with the product at the end of the day. And obviously, if you have the ability to up keep that internally, that’s great, and if not, then we can look at a further contract as well down the road.

10:20:09 AM - Speaker B

Right, okay, got it, thanks.

10:20:11 AM - Speaker A

Yeah. In terms of other customers that we’ve worked with as an organization, Terry, I want to be completely honest, we don’t do a fantastic job of capturing the customer success stories. What I can tell you is that we’ve worked with a number of different financial institutions and we actually have an entire vertical internally for financial services. We are working with, I can’t name the name, but one of the largest banks in Canada to date. We work with a ton of fintech startups as well as specifically in cryptocurrencies as well. So a lot of experience on the delivery side working with, I guess fintech to some capacity, but really just the financial companies altogether. Right. Okay. We do have a Success Stories web page on our site, but just flipping through it here, I don’t see any. And honestly, I think that’s designed for a reason and that’s because as I’m sure you’re aware in the financial services space, not a lot of these companies want to expose some of their pitfall to their competitors right. Just knowing how competitive it is, so so engineering on demand, terry? We have two specific offerings that are kind of in the same bucket or realm of offerings within the offscreen ecosystem. The first is called CRE or customer reliability Engineering. This is essentially to make sure that you’re andrew, I’m sure you can add a couple of points of context here as well, but CRE is essentially to ensure your product and services up. Right? So it’s 24 x seven by 365 uptime monitoring there’s incident and response team. So essentially what happens is if there was an incident within your environment, our team would go in again twenty four seven to sixty five and really just didn’t get that back up and running for you as soon as possible. That is definitely retroactive, whereas engineering on demand is proactive. Right. So the engineering on demand specifically is within office hours, so nine to five days a week type thing. But we’re actually working with your team to work through essentially a defined backlog that you might have for the engineering team. But it’s really around continuous improvement of your DevOps and infrastructure space specifically. Right. Andrew, was there anything you wanted to add to either one of those there?

10:22:30 AM - Speaker C

Yeah, I think the CRE component is great for smaller teams because you might have a subject matter expert on AWS on your team, but even that one person hopefully isn’t doing the 24 x seven coverage for the entire environment.

10:22:47 AM - Speaker A

Right.

10:22:47 AM - Speaker C

So you need to take into account on the spreadsheet, it’s two and a half engineers to cover a full 24 x seven, especially when it comes into vacation time and things like that.

10:22:59 AM - Speaker A

Right.

10:23:00 AM - Speaker C

So there’s definitely a broader headcount. So this works well for smaller teams that just need that 24 x seven coverage. And then I think this dovetails nicely with the EOD because sometimes when we’re getting alerts or any sort of findings from the CRE team, that’s the day to day operations, but they don’t necessarily have the capacity or capability to deploy enhancements to the system. So it might be, hey, CRE receives the alert, this is the remediation step and then that actually becomes part of the backlog for the EOD team to then go in and execute on further automation. That make sure that A that alarm never happens again, or B if it does happen again, that it’s remediated with automation as opposed to a ticket into the CRE team. So it’s nice to kind of work in tandem to monitor and then enhance your environment on an ongoing basis.

10:23:55 AM - Speaker B

Okay.

10:23:57 AM - Speaker A

Any questions? Terry?

10:23:59 AM - Speaker B

So far, so good. Thanks.

10:24:01 AM - Speaker A

Okay, great. Yeah. So honestly, I think that’s pretty much it in terms of what we are planning on presenting to you. I would love to understand how this stacks up with either a current vendor that you’re working with for the infrastructure and DevOps side of things, or really just if there’s any interest in seeing a statement of work from us just to get an understanding of what it would entail.

10:24:25 AM - Speaker B

Sure. We haven’t been looked into outsourcing the infrastructure part because we do have our own alert system right now, and it’s quite intense sometimes. So our own engineering work often got cut off due to bug fixing or dealing with alert from the server, AWS server. So these are the things that we might want to take a deeper look if we can outsource it. So I guess I’ll have to talk to the team and yeah, like I said, the rate is kind of important further down the road. As we are exploring this need to outsource, we are looking at the service levels and also the rate as well.

10:25:19 AM - Speaker A

Yeah, no, it totally makes sense. And so for Siri, I’m sure you can imagine that because it’s only monitoring for uptime specifically, there’s really no AWS funding that offsets that. So we can look at pricing for that specifically, but for engineering on demand, you’re looking at for every 20 hours block of time, you’re looking at about Cad5000. And so what we have right now is actually an offer for the first three months. We would basically leverage AWS funding and get 60 hours, obviously, for the course of three months, 20 hours a month, knocked down to roughly $2,500 Canadian for you. Right, so 60 engineering hours for$ 2,500 roughly. Right, okay. So incredible value. But I think the goal, and a lot of organizations don’t necessarily want to go through the time of onboarding a third party DevOps firm to then link public three months later. Right. So a lot of organizations are using this three month trial. It’s in quotations, a trial period, for example. So they pay roughly 2500 out of pocket. They get three months worth of service from us, and then at the end of the day, they get to decide whether or not they want to extend that engine on demand effort to a year contract or whatnot. I can tell you, the longer, obviously, like anything else in this world, the longer you sign up for and the more hours you sign up for, the lower the hourly cost is, because it’s just obviously you’re buying in bulk at that point. But we have some customers start this off at 20 hours a month and they realize, holy crap, we’re never going to meet our deadline. So they actually want to bulk this up to 100 hours or 200 hours every month. Right. So there’s essentially unlimited scale. And then obviously we’d love the opportunity to go forward with this at some point in time, but also just want to get your thoughts and feedback on that price point, obviously, the thought process, the time, all of that stuff as well.

10:27:08 AM - Speaker B

Right, yeah. Not a lot of idea on that. Point. So we are hiring people to do our own development work, but those are kind of different because they are highly specialized. We basically got no choice. Whoever we like to work with are just one guy in the whole open source world that we could find. So they’ll just name a price and then we’ll pay for it. So that has been how things goes. But they are reasonably priced, I guess, sometimes quite surprisingly. But anyway, we don’t have actual experience with these kind of outsourcing yet, so I’ll try to get back to you on that.

10:28:03 AM - Speaker A

Yeah, no rush there. I mean, again, we’re at no pressure organization whatsoever. It would provide you as much information as you need and we’ll obviously let you make the decision for your company and organization. But I guess the one thing that a lot of our startups think about as well when they do move forward with engineering on demand is that DevOps and infrastructure work is relatively mundane and boring in quotations. Sorry if I’m offending this call, but it is, it sounds like you guys are doing some really cool forward thinking AI ML stuff. And so the thought process is, well, instead of running all of these DevOps this work in house, why not just outsource the boring stuff and focus on the innovative creative stuff that’s really going to drive the needle forward for your company, right?

10:28:45 AM - Speaker B

Yeah, we did talk about internally how to better have a framework to get everyone’s work easier. We’ve talked about it, but we haven’t got to the system design point yet. So perhaps that’s something we would like to take advantage of the special deals that you mentioned to start from there, anything that you may add?

10:29:22 AM - Speaker A

The only other thing that I would add around entering on demand specifically is because they are cloud engineers, they’re not architects. For example, we would need a clearly defined backlog of items for the team to work through or remediate specifically. But if there isn’t that clearly defined backlog, there are other engagements that we could work together towards to create that backlog specifically. But then obviously it would kind of take a step back. It would kind of dip in the same funding program. So if we went forward with A, you couldn’t necessarily use the same funding program for B and vice versa. They’re almost one of the same in terms of the funding pool. There’s options essentially to help you build that backlog. If it was something you were interested in, you didn’t have that. Right. So I just wanted to leave you with that point as well. Okay. But yeah, that’s pretty much it.

10:30:08 AM - Speaker B

Okay. Regarding the specials and the quote, will there be a follow up email that I can share with my team or I just have to summarize myself and to share it with them?

10:30:20 AM - Speaker A

Yeah, we’ll send you an email. Terry, just to summarize the engineering on demand specifically, was there any interest in the customer reliability engineering or really just focused on engineering on demand.

10:30:31 AM - Speaker B

A lot of the terms, I’m not familiar with what they entails. So what I’m going to do is that so I’ll take your quote and rate to see it as something that we can afford to try. And then we’ll formulate our own needs and share them with you so that you will know what things we want to do and which buckets that may fit into the picture of yours.

10:31:06 AM - Speaker A

That’d be great, Terry. And so we’ll summarize the two different offerings that we spoke about here today. But at the end of the day, if you send us what’s important to you specifically, there’s a number of different ways that we can meet you where you are in terms of whether you need a data specialist or whether you need software architect or whatever it might be, you let us know and we can get creative in terms of how we deliver for you and obviously, everything we do. If we can, we leverage AWS funding to really knock down the cost for our End customers.

10:31:33 AM - Speaker B

Right, cool. Yeah. Thanks for your time. Scott and Rahim. All right.

10:31:43 AM - Speaker A

Yeah, no, likewise. Thank you so much for the time today, but very nice to meet you and looking forward to touching base again in the future. And we’ll send you a follow up email with, again, just a quick outline of CRE and EOD specifically and give us a couple of price points as well for those and just kind of summarize everything we chat about here today.

10:31:59 AM - Speaker B

Okay, wonderful. Thank you very much.

10:32:01 AM - Speaker A

Thank you so much, Jeff. You have a great day.

10:32:03 AM - Speaker B

Bye.

10:32:03 AM - Speaker A

Thanks. Take care