Ep. #141, Postgres as a Platform with Vignesh Ravichandran of Omnigres

Brian Douglas: Welcome to another installment of Jamstack Radio. On the line we've got Vignesh. Vignesh, how are you doing?

Vignesh Ravichandran: Doing well, thanks, Brian. Thanks for having me.

Brian: Yeah, pleasure, pleasure. Thanks for joining me here in the East Bay. We're still remote, but you're one of the rare occurrences where I have a guest who's on the side of the Bay that I'm also on, so I appreciate you taking the time out of this rainy morning. Real quick, did you want to introduce yourself and tell us what you do and why you're here?

Vignesh: Sure, yeah. Happy to do so, and before that, Happy New Year.

Brian: Likewise.

Vignesh: Myself is Vignesh, the co-founder and CEO of Omnigres. Omnigres has been around for close to a year now, we incorporated late last year and before Omnigres I was with Cloudflare, working on their Postgres database platform team. A giant MS and Integer contributor very early, around 2019, one of their biggest requirements at that point is, "We need a rock solid Postgres as a platform."

Think of like an RDS, that was my job there. I spent almost four and a half years, pretty good team and also it was great timing joining them around 2019, the IPO and seeing how the company scaled as a public company for the last five years. I've been around database and specifically Postgres ecosystem as far as I could remember.

My first job was Oracle DBA and I guess now it's no longer relevant, or at least not that much relevant. I spent three years at Ticketmaster, scaling their database platform before pretty much any live events, NBA, NFL, all their live events and venues. So that's me in a nutshell.

Brian: Yeah. So you spent some time at Cloudflare, you mentioned there Postgres as a service. What was the name of that service? I think it's still around.

Vignesh: Basically, the service was only for internal teams.

Brian: Oh, it's internal. Okay, I wasn't sure if it was just that R3 tool. So this was for internal teams specifically?

Vignesh: Yeah, internal teams. The database that we built, managed and maintained is only used by internal teams. For example, our customers saw an API of Cloudflare.com, our dashboards, all the zone, user data, billing information, everything is stored inside our database. Yeah, not exposed to that as a service to external customers.

Brian: Okay. Sorry, I heard what you were working on and I was correlating this with something completely different. But that's intriguing, so you built this Postgres service internally. What was the reason for making the decision to build this internally, as opposed to pick another tool off the shelf?

Vignesh: Very good question. It goes back, I think, to two things that the company, Cloudflare, believes so strongly that anything if it's fundamental, that has to be an open source tool.

Pretty much all the fundamental pieces of the infrastructure or platform pieces of Cloudflare is built on top of open source software, or it's completely built in house from the foundation. They just build it.

For example, Linux Kernel, open source Clickhouse. Cloudflare is a very, very big user of Clickhouse, starting all the way back in 2017 or '18. Open source Postgres from day one, 2009 Cloudflare started with Postgres. There's no they started with Oracle or on SQL, et cetera, and then eventually migrate to an open source database. In that time, we have seen so many companies have been doing that in the past, but it's the other way around with Cloudflare.

They just started with open source. The second reason in all the subreasons for why they do that is the ability to inspect what the software is doing. At the end of the day, it's just a bunch of code. If things don't go well or work as expected, they want the freedom to go back and fix and modify. To support this point, they have an internal full of Linux kernel.

We maintain multiple patches for Postgres. A lot of them we did submit upstream, sometimes they do get accepted, sometimes they're so specific for our needs. But at the end of the day, we have the ability to modify things. Another piece of software called PGBouncer, a connection which is very powerful in the Postgres ecosystem, we maintain a fork of PGBouncer because we submitted so many patches of it. It's a very long conversation and whatnot.

We ended up deciding, okay, we'll still open source it, maybe others will get benefit. So that's the main reasons why Cloudflare maintains.

Brian: Yeah. It also shows a sign of maturity for the engineering team to be able to have that flexibility to, one, maintain some of this stuff in the open and also give some of this back upstream. I spent a lot of time at GitHub and at GitHub they maintain a fork of Ruby on Rails because that was the decision that was made to build the original product.

They made a very clear decision back in 2019, I believe, is when they got back on the main branch of Rails. So rather than operating a fork, GitHub was able to get back on the mainstream of the Rails community. But that spent a lot of time of back and forth in conversations, and having core team members working at GitHub to help in that conversation.

It's pretty awesome to hear that Cloudflare has a really good, strong open source culture because it's something that I'm deciding between Cloudflare and something else. I know Cloudflare works pretty closely to the upstream of what they're owning. It gives me solace of I'm making a right decision technically for my product, knowing that they won't diverge into a weird enterprise.

I think you mentioned Oracle DBA, Oracle historically being problematic in that situation of needing to use Oracle specific things and then getting stuck into a weird contract. Definitely it gives me little bit of concern when I hear Oracle nowadays, but I'm very excited about things like Cloudflare. I did want to talk about Omnigres, so how did this come to be? How did you get started working on this?

Vignesh: Yeah. It all goes back, Yuri my co founder was the author of Omnigres. He has been hacking on this project for almost 12 months. The project was trending on GitHub and he was tweeting a lot about the challenges that he has been running into and the things that he was accomplishing and whatnot. That definitely caught my attention because I was already in the ecosystem of Postgres and what he was doing with Omnigres was very unique at that point.

So we both connected, this is somewhere around 2023, middle of 2023 and then we just built on for a period of a month or two, and we figured out there is something to be done here. He already had the pieces at that point in time, a good foundation and a platform to showcase what can be done with Postgres. Then me joining the team, able to go full time on that.

Around 2023 Fall we decided, "Let's do this thing," and that's how Omnigres got incorporated and started, and got some good investors immediately backing us. That gave us a bit of validation on what we are onto is something very interesting. I've just got two offers for more people. So yes, that's the genesis story of Omnigres, the idea that we can make Postgres into an application runtime.

Most of what we have heard of conventional ideas, is that Postgres on pretty much any database is just a database. You put data and get data. Especially the last 10 or 15 years, there's been more push that you should just use your database as a database. The approach of Omnigres is a little bit tangential to that, saying, "Why not use your Postgres which has a very good database? But can you also use it as your application runtime?"

So that's the initial idea, and we are still figuring a few things around what's the relation GDM is going to look like? What are the specific use cases? And audience we are going to capture.

Brian: Yeah. Curious to find out what brought you to build Omnigres?

Vignesh: That's an interesting question, especially the Postgres ecosystem if you think about it for the last two or three years, there are so many startups and so many interesting projects also popped up. Which is very good, also in a sense it's also not so good because, well, now you need to stand outside of the crowd. One of the biggest questions we had with really good investors is that, "What are you going to stand out of the lines?"

Anyway, I will tie this back to the answer of why I brought that up, because personally when I saw Omnigres, it was so unique. A lot of companies were doing things, innovative stuff around Postgres as one package in Postgres. It's like Postgres itself is notoriously hard. That's what I did at Cloudflare, making it production ready.

Postgres is powerful, no doubt on that. But making a solid and robust platform, a production ready one which has good observability, good backups and restoration, good security, all of them needs a good team.

And there's a lot of open source projects out there. Assembling that itself is a very good show of how big the market for that is. Then there is obviously several components that have been peeking out, et cetera. The idea with Omnigres was that none of those, it wasn't really packets solving the packaging issues, nor the serverless in order to sync that component, it's more of addressing the interior complexity or election in complexity in the software stack. Over the last few years, I'm not sure if you have noticed, but even building somewhat non-trivial data intensive application needs a lot of components.

Obviously you need a rock-solid database, then you need... most of them are using microservices, backends written in multiple languages. Now you need an orchestrator to manage all of these microservices, then pretty much talking to a database is slow. Then we pick some kind of cache like Memcache or Redis and whatnot. Then all of these microservices we ruled, all of them need to communicate someway and that's where we go for job queues like RabbitMQ or Kafka, some kind of queues.

This just becomes like a zoo, what we call a tech zoo, to build something somewhat non-trivial needs all of these components. The approach of Omnigres is taking a step back and asking, hey, do you really need all of these components? Is that the only way to build your data intensive application or can we simplify that? And the answer is that, yes, we can simplify.

Obviously one approach is coming up with a new paradigm, like a new programming language or a new kind of magic storage engine. But the challenge is building something is so much resource intensive, it's going to be 10 to 15 years. Even if you can build, let's say you have a lot of developers who can build that kind of a massive technology, the next question obviously comes in, how are you going to get adoption?

That's when we realized that building on something that already exists, and then can we not go along with that? Go for a ride, take them for a journey. That's when we looked that actually Postgres had a lot of these in terms of more as a capability. You can really, really do that. But the 99% percentage of them don't do that, because why do that?

That development, that debugging and the deployment experience is so bad that people don't even try that. We had conversations with our customers, they are like, "I will do that only if I really, really, really, really need to do that or use that." Well, okay. So that showed us that there's a problem here with respect with D3X, development, debugging and deployment experience.

Omnigres started solving that. I saw those beams of what the universe is able to do with just 1%, then that made me think, "Okay. You know what? We can actually do much bigger things with Omnigres."

Brian: That's cool. You mentioned the VCs asking about this crowded space and how you stand out, and that's a great answer too, as well, by the way. I'm getting similar vibes. Back in 2020 I had a conversation with Superbase back when they were still three people, exactly the same size that you are today. What was interesting about that conversation, but also looking at it now, is it's like I think they got a little bit of a headstart on leveraging Postgres the platform to solve some problems for enterprise stuff.

It sounds like there's a lot of folks who are doing Postgres as a service, how does Omnigres fit within that mix today? Are you providing the full-on platform like an app write, or are you providing closer to Postgres in extensions? I'm honestly curious about how I would leverage this.

Vignesh: Yeah. I think it's everyone is curious, and we chat with them. So it's more of the latter. What we have built so far is basically a constellation of extensions. If you go to Omnigres GitHub, you will see people get inundated sometimes, like, "Wow, this is a lot of extensions." So we are now close to 10 to 15 extensions built that all augments a vanilla Postgres, meaning that as long as you can install extensions on your Postgres database, you can start using Omnigres.

Whether it's Superbase or Tembo or whatever vanilla Postgres that you are maintaining on your one data center, it doesn't matter. You can just start using them. In the sense that Omnigres coexists with a lot of these platforms as a service providers, we see them all augmenting their Postgres offering, instead of we're trying to be yet another Postgres platform.

Brian: Okay. Cool. So appreciate you explaining that too, as well. It's always fascinating because I think the Postgres extension ecosystem, it's very vibrant. Postgres is definitely, I believe, the right choice. I know there's other choices for databases out there, but I think the community, the open source community that comes along for the ride helps get you started even more, even quicker.

Especially if you have that mental model of like, "Okay. Postgres starts here. Postgres extensions start here." So what Omnigres is basically packaging is extensions in a way that you can get started much quicker. I guess my question to you is what's the use case? Who is the ideal user for Omnigres today?

Vignesh: Yeah. So the ideal users are, I would say, they're SMBs who have already found product-market fits and they are already building. Developer size, let's say 10 plus developers who are building a backend application using Python or JavaScript or any language and using Postgres as their database of choice. In terms of the domain, most of what we speak to resonates well with fintech and eCommerce companies, in general.

I mean, I like to say anywhere money is involved. The reason behind the why these customers resonate much closer with what they're building is they care much more about latency, like eCommerce companies, their checkout flow needs to be super fast. So it does for fintech companies like Robin Hood and whatnot. They also care about consistency. It's not like they can just fire and forget or YOLO.

No, this is all... People get upset. If you say that you are going to send something, ship, and then if you don't ship if you're an eCommerce company, at the end of the day someone is going to be unhappy which means that you care about consistency as a property. These are people that already think Postgres, and for them the bigger challenge is that, "Oh well, we now also need to find the complexity aspect."

When we pitch them that actually you don't really need to find all the complexity. You already have your Postgres database where you store your data, why not build your application on top of it? These customers so far have resonated very nicely with our idea.

Brian: Cool. So then from your website there's a line about collocating code and data. Can you talk more about that? Is this like the SDK or have the interaction of Omnigres?

Vignesh: Yeah. So that's a fundamental idea for Omnigres is that moving your logic, application or business logic, close to where the data is... It's like if you think about for any business, just code alone without any logic, it's pretty much useless. It's just a bunch of bits and bytes stored on a disk. What value do you get out from just the data?

Similarly, you have, let's say, a lot of code there and doing some kind of heavy computation but they all need data to get some meaningful data or output, something out of it. Then why do we keep separating that? From the end of the day, holistically, we need these two things to generate some business value. That's one of the fundamental ideas of Omnigres, is to enable people to push or to put their logic close to where the data lives.

In very simple terms, this is nothing super novel if you think about it. That's really the way how applications 20 years ago were built. Anyone used to Oracle or a SQL server, they're all familiar with this approach. They have used PLSQ a lot, TSQ, these kind of procedural languages that let's you write your business logic. Very complex business logic.

These are not like 10 or 100 lines of code. People have written 10,000 lines of procedural language, that's pretty much how the bigger banks today, the insurance companies, even as off-grid that's how they operate.

Brian: Yeah. I guess my question to you, because you mentioned this in passing, that if you're using Postgres you could use Omnigres, or if you could install Postgres extensions... Specifically what you said is that you could use Omnigres. Because there's a lot of companies still building software like we were 20 years ago, but then there's a ton of companies who are not. I guess my question is the aim for larger enterprises that fit that larger model or my new SMB or my new startup can also start with Omnigres as well?

Vignesh: Actually we are seeing from both sides, the newer startups which are building. For example, our first customer, they are building their entire platform or their entire backend using Omnigres. They are building like Robin Hood for private equities, like a fintech company or I would say a fintech startup. Their entire application is built on top of Omnigres.

On the other hand, the bigger enterprises, since Omnigres as I said is a bunch of extensions, they don't need to buy in our innovation from day one. They are like, "Oh. We have this specific problem." For example, one thing that we have worked on is called OmniPython which makes Python as a first-class citizen inside Postgres. These are like enterprises, "Oh, we have this big logic returning to a procedure, like PLPGSQL. Can we just replace this one particular piece? Instead of SQL, we write that in Python using packages or libraries that Python provides?"

So we are seeing adoption from both ends. Enterprises, I would say, more like smaller pieces or smaller components they are trying to replace, and then early stage startups are using us like the whole-sum of Omnigres.

Brian: Yeah. That's fascinating, man. The embedding the Python inside of your Postgres, that's a lot of what I see in these new database as a service type tools, it's folks are bringing in a new paradigm or a new language to interact. But if you already have a strong experience in Python, I think that might be super enticing for folks to check out OmniPython.

So I guess my question to you, and I'm kind of winding down the conversation... We came into this conversation, before we hit record you were sharing that you're very new, early, y'all are not just getting started, but still on the cusp of how to figure out this product-market fit. What's next for Omnigres? What spaces are you looking to enter?

Vignesh: Yeah. So I mean, definitely trying to figure out this platform of, hey, we can solve or make, for example, Python as a first-class citizen which is very good. But more than that we're trying to figure out what specific use cases are problems that we can uniquely solve. For example, one thing that we are looking at is that providing audit logs as a service.

Think about any company, somewhat bigger companies, they all need to know when a record gets changed, who changed it, and what was the before image and what's the after image. Right now people have to go through some kind of tools like DBCM and DBCM needs Kafka which has this Change Data Capture or log streams, or logical replication it uses to find all of that.

At Omnigres we are like, "Well, how can we solve that? How do we approach this problem?" We are like, "Well, we are right inside your Postgres database, which means we know what the before image and after image. We can build this audit log as a service with just a single Postgres exchange."So that's just to give you an example of what we're working on.

Obviously the next one we are working on is a query cache. Think about as I said going back, talking to your database is slow. One approach is put your logic inside your database so that you are no longer talking to the database, you are talking within the database. Or even for whatever reason you still need to do that, let's assume, not everyone is going to change their application tomorrow, the way how it works. Why? Because it's been working for 20 years, so let's keep them.

So still they only use Redis or some kind of a key value store. Right now they have their relational database, while it'll be likely Postgres and they have Redis or some kind of a cache. When we looked at how can we take this component out of your tech zoo, we can build a query cache inside your Postgres. So there are early experiments that currently we are doing, so hopefully some of them will be materializing in the next, I would say, couple of months.

Brian: Excellent. Yeah, you had mentioned the GitHub repo, folks can pay attention to there. I don't know, it seems pretty active, y'all got over 1,000 stars so it seems like there's a bit of a community around it and a handful of issues. I'm a big fan of open source so I recommend everyone watch the repo, star the repo and you'll probably get a lot of context there. Any other places that you would recommend people to keep tabs on the story of Omnigres?

Vignesh: I would say one other is our Discord server. We have a Discord community where the contributors, Yuri, myself and Josh, we all are very active, and also our extended teams. We have contributors who have solved multiple bounties, we even have a bounty program which we created because our people are really interested and excited to contribute and we thought, well, actually let's also compensate them fairly. Contributing is good, but let's make it win-win for everyone involved. So if someone is interested in hacking our low-level stuff then, yeah, definitely take a look at our GitHub repository. It's all written in C, so good and bad.

Brian: Yeah. C, I always find it's super readable for me, surprisingly. But I love perusing through code I didn't have to write and learning something from it, so I encourage folks to take a deep dive, take a tour into the Omnigres repos. So with that, Vignesh, we actually want to transition to picks. I appreciate this conversation on Omnigres.

This is now an opportunity for us to talk about stuff that we're jamming on, so it could be music, could be food related, could be tech related. If you don't mind, I'll go first. I've got two picks for you today. I just watched a movie over the weekend and was actually really surprised by it. There's this thing that I've mentioned on this podcast a few times, me and my spouse, we always try to find a movie to watch if we have some time away from the kids or the kids are to bed on time.

But the challenge is trying to find something to watch, so we've been doing this thing where we use Amazon Prime which we pay for it, don't always stream stuff from there. But as of lately, usually around the awards season there's always movies we've never heard of or that have won awards at different Sundance or Indie film festivals and they always end up being on Amazon Prime for some reason.

So we kind of pick the first thing we see usually, and then we watch it and if we don't like it we'll switch to something else. But I guess if you didn't have to pay for a ticket, you already sunk cost into the streaming... I guess what I'm getting at is the pick is the movie called Another Version Of You. It was at a film festival in Tennessee, I guess there's one out there that it won out there. It's about a guy who gets gifted a key to unlock a door to a parallel universe, and the one caveat is that you can never go back.

So he finds himself in a situation where he's at the point of his life where going to a parallel universe is actually a little more opportunistic than where he is today. Like any sort of sci-fi or this whole conversation around parallel universes where it's binary, either you do this or you do that, or you're with this person or that person, or you went to this school or that school.

Every different universe is a fork in some decision that he made in his life. So he goes through the world, basically chasing a girl, and trying to find a situation where he can be with a girl he's looking for. So I'll leave it there. It's not high budget, so it's decent acting, not the best acting. But I would say for a movie that would have you think about the movie after the fact, that's the one thing I appreciate about it.

The movie is called Another Version Of You. I do recommend it. If you're in the States at least, it's in Amazon Prime I their streaming side of their product. I assume you probably pay for Prime because everyone has Prime at this point.

Vignesh: At least I do. I was joking. I guess Amazon Prime and Costco are the two things that's just going to be by default, every household in America is going to have it. It's like your SSN, you pay in and you're going to have those two.

Brian: Yeah. So Amazon, Costco and GitHub, everyone's going to have a GitHub account as well. Yeah, so the other thing I want to mention in passing because I've been doing a ton of data visualizations, so we do use Postgres internally at OpenSauce and I've been messing around with Neevo.rocks. Neevo is like a library on top of D3 and honestly I'm not even sure if D3 is still a part of one of the dependencies of the project anymore.

But if you want to visualize data and React components, Neevo has a bunch of really good interactions and charts and an array of different ways to display data. The game I've been playing is I've got a bunch of jSONs or a bunch of stuff coming from SQL, how do I present this in a way that's visually appealing inside of an interface?

So I've been messing around with this site called Contributor.info where you can just go to and by the time this podcast comes out it'll be much nicer, but today you just add in any repo and you can see your contributor info. So looking forward to making this more of a shareable experience in the future, but it was a project I started working on the last week of the year, right after Christmas.

I tried to build something, like I did my new website or blog. This happened to be I'm building a little side project for myself to display OpenSauce data.

Vignesh: Yeah. I just put in, obviously, Omnigres just for people who are going to try it. It looks like they have to put the project name and as well as the user name or something.

Brian: The repo, yeah.

Vignesh: Omnigres/Omnigres.

Brian: Correct, yeah. In GitHub world it's called the full name for any project, so it has to be the owner and the repo. It's not spelt out in the app, but once I get it working, the data working in Neevo, I will eventually have a better experience so you don't have to figure that out.

Vignesh: Always happy to be an early user, right? So my things are... Definitely nice suggestions, Brian, so I will definitely look out for the movie and it's something we, with my spouse, I run into often, that same challenge. My picks are, the very first one is something that I use every single day, it's my espresso machine. It's called a Gaggia Classic.

Again, my spouse gifted it for I think one of my birthdays, two or three years ago. From that day onwards, I've been using it every single day. I'm a coffee lover, obviously, or whatever, liker. But the thing is I like the coffee in a very specific format or way. I don't drink black coffee. The way how I drink is just one shot of espresso and then put like 80% of milk in it, steamed milk.

So it's kind of like off-white, what some people call a full-white, but that's the only way I like it and not a lot of coffee shops make it that way. I tried so many times, Starbucks and whatnot, they just go with their usual way how they make it. Maybe the smaller cafes I've tried, they do make really good ones, nice ones. But the challenge has been finding something in the East Bay side, I have to travel like 20 minutes one way and go somewhere, and the lottery here and there.

Another funny stuff is that a lot of coffee shops also closes at two o'clock or three o'clock, so I'm actually getting somewhat relaxed and I can drink actually a coffee and enjoy it after 3:30. Well, guess what? Already shops are closed. So my wife, Mina, obviously looked at it and understood and she gifted that. So that's something I really, really recommend anyone who's into coffee.

The other one, since we are talking about coffee, is the beans. That's the next one. You have an espresso machine, so how are you going to make your coffee? I've tried so many things, the Bluebottles and what not, Sideglass is the other one which I tried. But my pick is Whirl. This is a roaster based out of Santa Cruz, amazing, amazing. They have a subscription model. I wouldn't say that they are the cheapest one.

Obviously if you go to Costco or Walmart, you may find something less expensive. But this is the best. This is worth that extra $5 or $10 that you put into your coffee. That's my food related pick. Let me think about more on life related. There is one book which really changed the way how I perceived the world or see the world. It's this book called Give and Take by Professor Adam Grant.

He's a professor from, I believe, University of Pennsylvania or Wharton College of Business. Amazing book. It goes deep into this idea of how helping others eventually helps you succeed in your life. It's kind of like if you think about it, well, if someone is asking you something, the first thing you think is, "Why should I help? What's there in it for me?" But then this book is full of examples of how givers takeover the world.

He puts people into three buckets, givers, takers and then someone trying to match us. Obviously most of the world is matchers, because "Oh, you did something for me? Well, I will do something for you." But then there is a very small subset of people who does help without expectation, without nothing in return.

"I'm just going to give you, that's it. If you help me somewhere down the line, so be it. If not, don't worry, don't bother. You just keeping paving the will for other people, pass it on to others." So really that book really resonated for me. Yeah, definitely I recommend any books from Adam Grant and specifically Give and Take.

Brian: Okay, yeah. Adam Grant is somebody I have definitely seen speak at places. I think he actually was a guest speaker at one of the internal conferences at GitHub. I've never read any of his books or have any sort of background on who he was. I'm very intrigued by this book, so I just added it to my wishlist as you were talking.

Vignesh: Yeah. He showed examples all the way from leaders, people in politics and VCs. Definitely the very first example he gives is a VC who some people ran into and how they helped each other and ended up investing in some of their companies. But he gives examples from all over the place. Then once you get that idea, then if you look back even in your organizations, when some people gets promoted you're genuinely happy for them, you're like, "Yes, that person got promoted. They deserve it."

Why? Because they've helped you or you have seen them go above and beyond to help other people. So yeah, that's one book that I would recommend highly. The other one is Forest App. This is an app that I use pretty much every time when I try to get into work because obviously we are in the world of distraction, so I do get distracted time and time, and I want to keep my focus when I'm working on something or even I need to get something done.

This little app that they have a mobile app, you can install it in your i-wares or iPad. The best part is it's also a browser plugin or extension so you can enable that plugin and then it just makes you stay focused. It's like a small tree that you glue once you focus for that specific time, let's say 30 minutes you block and you want to do this thing, and then at the end of 30 minutes you see a tree popping up.

If you couldn't do that and you, for whatever reason, you went to Twitter.com or somewhere, then that tree gets dead. So obviously you don't want your trees to go dead, so focus.

Brian: Excellent. Yeah, sometimes we need a little nudge to not kill trees so we can stay focused and get some work done.

Vignesh: It's a lot more nice. I think staying on the track is such a... I mean, at least for me personally it has been a challenge. I see a lot of other similar authors and books that are saying how to focus in this distracted world, so any kind of help externally, internally that if you can get, well, go for it.

Brian: Excellent. Well, Vignesh, thank you so much for coming on the chat, on the podcast about Omnigres and sharing some pretty delightful picks, actually. I look forward to digging into these books and also I might even check out this focus app as well. But with that, folks, keep spreading the jam.