Elevating Client Onboarding With AI Agents With Rabih Ramadi
- Quik! News Team
- 3 days ago
- 29 min read
Rabih Ramadi is the Co-Founder and CEO of Avantos, an AI-native operating system that streamlines customer onboarding and servicing for financial institutions. Rabih previously served as a founding team member and Chief Revenue Officer at Unqork and was a Senior Partner at KPMG, where he led the Capital Markets sector. He holds an MBA from The Wharton School and an ME in engineering from Cornell University. Rabih also sits on the board of Leathwaite as a non-executive director.
Here’s a glimpse of what you’ll learn:
[2:06] Rabih Ramadi discusses Avantos as an AI operating system for client management
[3:56] Using LLMs and data modeling to build AI agents
[6:47] How ChatGPT’s rise enabled Avantos to scale solutions
[8:17] Three types of AI agents: frontline, operations, and product
[11:00] Rabih talks about automating advisor productivity with client insights
[13:30] Embedding compliance controls and human approvals
[15:00] Prefilling client data from CRMs, Plaid, ADP, and IRS integrations
[19:53] Managing risks of too many AI recommendations
[21:55] Avantos as a flexible client onboarding and servicing platform
In this episode…
Many financial institutions struggle with slow, manual onboarding and servicing processes that hamper efficiency. Advisors and operations teams spend significant time re-entering data and managing compliance, creating obstacles to optimal client service. How can financial firms leverage AI to streamline these workflows and elevate client experiences?
Rabih Ramadi, an expert in financial technology and artificial intelligence, shares how his team built an AI-driven operating system to address these challenges. He explains how linking clients, products, and advisors through AI agents helps automate tasks, prefill data, and trigger compliant workflows. Rabih emphasizes building flexible systems, embedding human approvals, and prioritizing the right recommendations so advisors can focus on high-value client relationships while still maintaining trust and compliance.
In this episode of The Customer Wins, Richard Walker interviews Rabih Ramadi, Co-Founder and CEO of Avantos, about using AI to transform client onboarding and servicing. Rabih discusses the rise of AI agents, how to prefill data from trusted integrations, and why compliance must remain central to innovation. He also explores the risks of too many recommendations, the future of chat-based interfaces, and the importance of pairing business knowledge with technical vision.
Resources Mentioned in this episode
Quotable Moments:
“Our agents are not replacing humans per se, but their job is to make the humans way more efficient and productive.”
"When we saw ChatGPT and OpenAI, we realized now that we have the right components and foundation to do it at scale.”
“We launched our first agent maybe four or five months ago. Okay. Yeah. For us.”
“The engineering mindset will never go away. Yeah, I think AI is going to do a lot.”
"We’re not forcing people to change everything they do, but our premise is, rather than doing it in an analog way, just doing it pure digital."
Action Steps:
Adopt AI-driven client onboarding tools: Automating data collection and account setup reduces errors, speeds service delivery, and frees advisors to focus on relationships.
Embed compliance into every workflow: Integrating audit trails, approvals, and validations ensures regulatory requirements are met while maintaining efficiency and client trust.
Leverage AI agents for frontline productivity: Intelligent agents surface insights, recommend actions, and execute workflows to boost advisor efficiency and personalize service.
Prioritize and filter AI recommendations: Curating and ranking actions prevents overwhelm, helping teams focus on what matters for retention and growth.
Prepare for chat-based interfaces: Conversational AI enables natural interactions, making adoption easier, boosting engagement, and future-proofing digital client experiences.
Sponsor for this episode...
This is brought to you by Quik!
At Quik!, we provide forms automation and management solutions for companies seeking to maximize their potential productivity.
Using our FormXtract API, you can submit your completed forms and get clean, context-rich data that is 99.9% accurate.
Our vision is to become the leading forms automation company by making paperwork the easiest part of every transaction.
Meanwhile, our mission is to help the top firms in the financial industry raise their bottom line by streamlining the customer experience with automated, convenient solutions.
Go to www.quickforms.com to learn more, or contact us with questions at support@quikforms.com.
Episode Transcript:
Intro: 00:02
Welcome to The Customer Wins podcast, where business leaders discuss their secrets and techniques for helping their customers succeed and, in turn, grow their business.
Richard Walker: 00:16
Hi, I'm Rich Walker, the host of The Customer Wins, where I talk to business leaders about how they help their customers win and how their focus on customer experience leads to growth. Some of my past guests have included David Steel of One WealthAdvisors, Cameron Howe of Investipal, and Joshua Rogers of Arete Wealth. Today, I'm speaking with Rabih Ramadi of Avantos, and today's episode is brought to you by Quik!, the leader in enterprise forms processing. When your business relies upon processing forms, don't waste your team's valuable time manually reviewing the forms. Instead, get Quik! using our Form Xtract API.
Simply submit your completed forms and get back clean, context-rich data that reduces manual reviews to only one out of a thousand submissions. Visit quickforms.com to get started. All right. I'm really excited to talk to Rabih, who's actually one of our newest resale partners here at Quik!
Rabih. Rabih is the co-founder and co-CEO of Avantos. Before co-founding Avantos, he was a founding team member and chief revenue officer at Unqork. Prior to that, Rabih was a senior partner at KPMG, where he led the capital markets sector. He holds an MBA from the Wharton School of the University of Pennsylvania.
And I guess this is a master's in engineering from Cornell University. Rabih, welcome to the Customer Wins.
Rabih Ramadi: 01:44
Thanks, Rich. Pleasure to be here.
Richard Walker: 01:47
It's awesome to see you today. So for those who haven't heard my podcast, I love to talk to business leaders about what they're doing to help their customers win, how they built and deliver a great customer experience, and the challenges to growing their own company. Now, Rabih, I know you're doing amazing things, but let's understand your business a little bit better. How does your company help people?
Rabih Ramadi: 02:06
So Rich. We at Avantos. What we did was build an operating system, an AI operating system for client management. The way we define client management is the way you onboard the client to a financial institution, and then provide the ongoing service to the client, all digital or through AI. That's the product we built, and it adds tons of efficiency to our clients and allows them to get clients quicker and with a better client experience.
Richard Walker: 02:35
All right. We're just going to dive under the hood here. Because when you say it's built with AI, why should the user care about that? Why should the customer care about that? What does that really translate to?
Rabih Ramadi: 02:46
It basically brings the capability through AI architecture to link the client, the client hierarchy, the client household to the product that they are buying from the financial institutions, and the agent servicing the client, whether it's an advisor, a broker, or an operations person. So by linking these entities, we provide the foundation to substantially automate the way you onboard the client or service a client with AI native agents that help through that onboarding process, whether to prefill data or to prefill the quick forms, or to do an integration with the portfolio management system.
Richard Walker: 03:26
Okay. So you have agents running in the background, is what I heard.
Rabih Ramadi: 03:30
That's correct.
Richard Walker: 03:31
How to go back in time. I mean, AI is a big term. Most people think of it purely as ChatGPT or LLMS, but it's a very encompassing term that's been going on for decades. What are you using? Like, is this all LLM-based?
I mean, is this machine learning? Are you doing computer vision? What kind of stuff are you guys into? Because I'm going to ask you next about how you came to the conclusion to build agents.
Rabih Ramadi: 03:55
So yeah, we have.
Rabih Ramadi: 03:56
An architecture, we have LLM, we also have other types of AI tooling, but we mainly rely on LLMs as well as some old-school data modeling capabilities to basically create these agents and embed them as part of the business, rather than creating agents for the sake of just having AI agents. We are trying to solve a specific business problem with the agents. Our agents are not replacing humans per se, but their job is to make the humans, whether it's an advisor or a broker, or operations, way more efficient and productive in doing their job, and give them signals based on data on how to better service their clients.
Richard Walker: 04:37
Does this mean in your mind that the agent can figure things out within your software, that the humans don't have to think through? Or does it mean, and or does it mean you could build your software more flexibly without such rigid structure and architecture too. I mean, I just don't know. This sounds amazing. It sounds like you got a bunch of minions running around.
Rabih Ramadi: 04:57
It starts with the building, which is like the integrations. If I had to integrate with the system quickly, I would have AI agents helping me to do these integrations. But then when it comes to the day-to-day servicing of a client, rather than relying on humans remembering what to do to a client, the software will still allow you to manually trigger workflows, but the AI is listening. The AI agent is listening to calls, emails, data feeds, and based on that, they are recommending all the additional servicing you should do for the client. And also, most importantly, trigger the workflow and the orchestration.
So we don't stop by recommending actions, but we go all the way to the execution of these actions, including the integrations with quick, as well as with portfolio management systems, CRM, and other systems.
Richard Walker: 05:46
Man, how long ago did you start this?
Rabih Ramadi: 05:51
We officially started the company a year ago, but what we did before we even started the company, we launched the V1 of the product with a big wealth manager. So we had all the time to really tweak the product and perfect it. And then we launched the market a year ago. And so far, knock on wood, the traction is amazing.
Richard Walker: 06:11
So prior to launching a year ago, how much time did you put into this to build this first version?
Rabih Ramadi: 06:17
Effectively another year, year and a half. But that's an idea that my co-founder and I have been thinking about probably for the last ten years. So this is a space that we know extremely well. Obviously, ten years ago there were no LMS or none of the data architecture that could enable us to build the product the way we built it. But that's the space we understand extremely well from prior experiences.
Richard Walker: 06:41
Did you feel a little bit like winning the lottery when ChatGPT finally made big splashes? Like, we can do it now?
Rabih Ramadi: 06:47
It felt something like this because it enabled us to know the crux of the issues. We knew the issues that made it super hard to scale in space. There's a reason why we don't think there's any solution that has been able to crack the space in a scalable way. So when we saw ChatGPT and OpenAI and all the new innovations and architectures, we realized now that we have the right components and foundation to do it. So we went ahead and built the company right after, because we realized that now we have the technical architecture capabilities to solve the problem at scale.
Richard Walker: 07:21
I feel the same way. Like for over a decade we've had the idea of how to get rid of forms. And I don't mean you can't get rid of the need to collect data. Your system needs it. Other systems need it.
But the idea that a human should have to look at what looks like a form, the image of a form with a small font. Hard to read this big a space where a big address. I mean, we've had this idea for a really long time, and we built a version many years ago that got us partway there. But to get all the way where we need to go, it's AI. I don't think we could do it without AI.
So when this started coming out and I started to realize we can execute on our vision finally, it got me more excited about my business than I've been in 20 years. I mean, it's so amazing what's possible. I want to ask a little bit more about agents, because agents are a fairly new term, maybe about a year old for most people. When did you guys click in with that concept of agents?
Rabih Ramadi: 08:17
We launched our first agent maybe 4 or 5 months ago. Okay. Yeah. For us, the concept of the agent is tied to the persona of the end user of the app in the real world. So we have three types of agents.
One, the first one is tied to the front line. So the advisor, the broker, the banker, they are mainly focused on the front office productivity. The next set of agents is more focused on operations. This is basically the agent helping the back office with its operations. Finance compliance.
Their day-to-day job includes a lot of unstructured Structured data, basically translation into structured, etc. and the third one is more focused on the product agent. So the product people, the product persona, the person designing the workflow or designing the app itself, how to make their life easier by having AI help them generate these workflows versus build them from scratch. This is how we think about agents.
Richard Walker: 09:19
No, I love this. And I think that's really appropriate. I want to talk about all three. But let's go back to the product. One.
First is that for your own internal team or is your product configurable by the customer. So they can say, I want it to do X, y, z and talk to the agents and make it happen.
Rabih Ramadi: 09:33
The key focus is the client building. Using our product, we have self-service. We don't want to be the one building all the orchestration. We come with preset orchestration, blueprints that we build based on industry best practices. But we realize every client, especially the larger clients, have their own ways of running a process.
So this is basically the tooling for them to automate these processes with AI agents that are smart enough to know how to generate the blueprint of the first floor. Obviously it still requires some manipulation afterwards, but at least the 70% can be generated through AI.
Richard Walker: 10:12
Yeah. That's amazing. So going to the front end, there is a lot of movement in our industry right now with all the note takers. There's 26 of them apparently. And I've had several of them on the show.
It's been fascinating to talk to them about what they're doing. There is a movement around how do I save the advisor time? I'd love to say I started that movement 23 years ago, because that's what I've been promoting this whole time, but I won't take credit for it. It's just that the world realizes when you give the advisor a better experience and more time with their clients, and therefore the client gets a better experience. You're driving revenue, you're driving loyalty, you're getting enhancements in the portfolio size that you get to manage.
What are you seeing out there? And part of this I wanted to ask is are you competing with these note takers because of that front end workflow, or are you complementary?
Rabih Ramadi: 11:00
We don't view them as a competition. We have a different angle, so we're very focused. One of our most primary focuses is the advisor's frontline productivity. Again it can be an advisor in wealth, a broker in insurance or a banker in the banking space. But the way we built the foundation of the system was to make that frontline persona super productive.
Now, the way we view it is by literally combining or by linking the data that the frontline needs to know about the client. So we're linking the client data with the products in a way, now that this frontline person have full access to the client, whether it's the householding structure of the client, the products or the services, all the ongoing engagements, all of that stuff is being captured by by our operating system, including meetings, by the way, meeting notes, etc.. And based on all these insights we're recommending to the front line agent is how to really automate the servicing of the clients. For example, I get an email from a client saying, look, I have a newborn in the family. So the AI agent is going to say, by the way, advisor or banker, the first thing you should do is recommend ending the newborn as a beneficiary to XYZ accounts.
In this case, we're going to take the data, execute the workflow, go quickly, generate the custodian form, and send it to the custodians. And number two, maybe you should suggest to the client to open a 5 to 9 plan and increase their life insurance policy. And once the client consents, then the triggering and the execution of all these events happen at Ventas. So it's not just the recommendation. The recommendation is interesting, but it's not enough from a productivity perspective because you still have to do all the work.
So we take care of all the way to the execution again and the integrations and anything that has a forum-based integration, whether it's the custodian or an insurance carrier. This is where we're leveraging the integrations with all the catalog of the 40,000 plus forms you have that cater for all the types of products and events that we need to generate to service these events.
Richard Walker: 13:14
Okay, so that leads me back to the operator agents. How are you making sure that these recommendations meet with compliance standards, that the communication back out to this client, if it's automated, is meeting compliance standards and appropriate language, all that kind of stuff.
Rabih Ramadi: 13:30
This is where my co-founder and I are, having spent a lot of time in regulated industry all our life. So for us, compliance comes first. Like so everything we do has a compliance component when it comes to audit trail, record, keeping, the approval chains. All this stuff is embedded in our platform natively, including things like notifications, etc. so all of these are pre-built capabilities, Including also all the data validations that were very fixed on to make sure we avoid niggles later. So all of these are pre-built controls embedded in the system.
We still require it, even though we can launch events that can run on its own through Ise with no approvals. We never allow for that for now because we don't think the industry is ready for that. So everything still requires a judgment, a human even, to just click on something to approve before it gets triggered because nobody wants to start opening accounts to clients without the clients basically approving them or before compliance approves them. So everything is embedded from that perspective, from a control perspective, while the automation happens behind the scenes.
Richard Walker: 14:38
This feels a lot like having, I don't know, a clone in a way of an advisor looking at the problems that are coming through these messages and saying, let's do this and let's suggest that, etc. this is fantastic. How do you collect data? Where is it coming from? Are you connecting with CRMs? Are you doing the intake process?
How are you getting all this stuff?
Rabih Ramadi: 14:59
Different ways like as you know, Rich and I appreciate one of the biggest challenges. When you onboard a new client, you service a client as it requires a lot of data. And currently the unfortunate thing in the industry, people type it from scratch. The good news is a lot of the data exists. It exists in different places.
CRM definitely is a good source of data, but there are other sources. So for example, from an account aggregation perspective, we have integrations with the likes of plaid and others where we can just basically link accounts, full accounts, pay stubs. We have integrations with the ADP and other providers where we can prefill data taxes. We can also integrate. We have integrations with providers who can give us ten years worth of tax returns from the IRS.
All of these are very key integrations for us, our biggest focus when it comes to data filling initially is to prefill as much as we can, whether it's from CRM and third party vendors. And the other concept is once the data is in events, the key thing for us is to eliminate any swivel sharing. So we're feeding them everything from CRM to the financial planning tools to the custodian to the portfolio management system. Our idea is like if you have the information, there's no reason on Earth to swivel, share it into other systems.
Richard Walker: 16:24
Yeah, man, it sounds like a lot of you. A lot of the work you've been doing is integrations. If you have all this work.
Rabih Ramadi: 16:30
We do a lot of integrations. That's why we built a very sophisticated integration layer. But then the key thing for us is anything we see as a repeatable integration, such as Salesforce, such as a custodian, we build connectors. We don't want to go into integrations. So these are connectors for us.
All this stuff we're doing with you. when it comes to the custodian integrations and soon the insurance carrier integrations. We do it once, then all our clients will benefit from these integration pipelines because, as you know, the integrations in financial services overall, some of these systems are old. So it's very expensive for every client to build these bespoke integrations. And that's why we build these connectors so all our clients can benefit from them at the same time.
Yeah.
Richard Walker: 17:19
So the last guest that I interviewed is Paul Osterberg of Security Base Camp. I think that's what I hope I didn't get that muddled up because I was thinking base camp security, I think security base camp. And he's in cybersecurity, he's a virtual CISO, and he's been addressing these types of problems for a long time. But our conversation went to the point of trust because you think about a cyber breach. Nobody cares if you never get breached.
They absolutely care when you do. And it's only a negative. You lost trust. So with that theme, I'm wondering how do you ensure you maintain trust with your clients because you choose the wrong CRM record to make a recommendation to somebody? Because the AI agent said, I figured this out for you and I'm 100% confident I'm right.
How do you stop that? How do you make sure that you're not losing trust?
Rabih Ramadi: 18:12
So 2 to 2 things from an event platform perspective, we have a lot of expertise in deploying secure systems. So the first thing we do is every client is a single tenant. So zero sharing between clients and the LMS we're using are private LMS. So we don't believe in tapping the public LMS pull data because obviously from a security point of view it's not great. So everything is private and very specific to the client with a private link to the environment.
So nothing traverses the internet from that perspective. So very secure. And to your point, the example you mentioned on the CRM record recommending something is the problem. The challenge with this is some of the data. As we all know, no, there's no concept of 100% clean data anywhere.
No. So you may have a bad drug in the CRM. That's why for us we have one internal number of validations before we recommend anything. But even when we recommend something, we also always for now have a human agree with it before it goes and executes. Because you're right, you can have a stale CRM record that was in Salesforce and Dynamics five six years ago, and the whole profile of the client changes. You don't want to base that on automatically executing something on behalf of the client without really having proper checkpoints.
So we always have the human in the loop verification before we trigger an event.
Richard Walker: 19:38
What do you think? Well, I'm curious what you think the risks are with this type of approach to the world. I like I'll throw one at you. Is there a risk that an advisor will oversell to a client? Because there's too many recommendations?
What do you think some of the risks are of approaching the world from this standpoint?
Rabih Ramadi: 19:53
The risk is when you have way too many recommendations. So we're very conscious of that risk. I don't think anybody has a magic bullet in terms of how to address it yet. But our biggest thing is we don't want to flood the system with tons of recommendations because, as you know, there is tons of data. Once you have a client, there's so much data on that client in different systems.
We bring it all together, which is amazing. But now you have a lot of data, which means you can recommend a lot of things. But if you recommend too many things, one, it dilutes the value of these recommendations. There is a risk of people triggering incorrect recommendations, or there is a risk of people ignoring these recommendations because they have too many of them. So we're working very hard to find the balance in terms of curating these recommendations.
And at the beginning, where they have less versus more recommendations. And as we as the LM become more and more mature, we can keep adding into it. And the second thing is the UI to surface it. You don't want to flood people with all these recommendations. There's a way within the experience of the app where you when I basically surface some of these recommendations based on trigger events versus, you know, having somebody's inbox flood with recommendations every day in the morning because it's not useful then yeah.
Richard Walker: 21:12
When my to-do list is 48 things long, I don't do any of them. Exactly.
Rabih Ramadi: 21:17
And also the concept of priorities, because you can have some small recommendations. You can have some crucial recommendations based on trigger events. So we're applying the concept of prioritization, which is like tell me the top three I should really be mindful of. Like there's a potential churn or there's a potential big opportunity for conversion versus something nice to have. So we're working on the UI and the experience to differentiate between these different levels of recommendations.
Richard Walker: 21:46
So I know you described this as the operating system for financial firms. Is this a new account opening system too? Is it primarily a new account opening system?
Rabih Ramadi: 21:55
It's an operating system for client management. So we're not claiming to be the operating system of everything. So onboarding and servicing this is the space we focus on. We're not focused on the selling piece by the way. So we assume the CRM is where you do the pipeline management. The lead gen posts basically the client telling you I want to become a client.
So the onboarding that includes things like the data gathering, the KYC, the account opening, the underwriting when it comes to insurance. And then once the person becomes a client, all the ongoing servicing events, everything from adding a beneficiary to changing an investment strategy to money movements, these are the things we're really focusing on. There are a lot of other things that fall beyond that, and we're not touching that.
Richard Walker: 22:43
Yeah. I think one of the challenges in our industry with services that offer new account opening is they get to a point where it's hard to change what they first built, what they first scope out takes a lot of effort, and it's hard to change afterwards. And it's expensive to change and maintain over time. Do you think you have that problem solved?
Rabih Ramadi: 23:04
We do. We have seen it. So the good thing is we learn a lot from what I've seen in the past. So we build the flexibility from a data foundation perspective so we can extend the products and the type of accounts we can open in the future. So for us, the the core foundational thing when it comes to account opening one, it has to support multiple account types, two multiple channels. It's multi custodian in the case of wealth.
And then beyond basically just the wealth products like if the client wants to basically if the wealth manager wants to offer insurance products, banking products, the client for us these are just products. Now the data model is different. In the case of insurance there is no custodian. as a policy administration system. But for us all these are different integrations.
But the data structure we created allows for us to incorporate all these different product types in the future. The biggest problem currently with account opening is it's very bespoke. It's done for an account type or for a custodian type. And that's super annoying for companies because you end up not really selling as many accounts as you want, or operationally. You have to do it one at a time, which is very not scalable in this case.
Richard Walker: 24:19
Oh yeah, we had a customer. I'll leave him nameless, but they had some 200 transaction types or account types, and they went live with the first 65 or so. And then it took three years to get the rest.
Rabih Ramadi: 24:31
And I can tell you, Rich, I have seen too, like some people end up doing the first 65 and then they give up on the others. Like the problem it becomes one when it takes too long, obviously super expensive, but then the ongoing maintenance because as we all know, these account types, they change. There are new types. The custodian changed some of the data structures. So unless you build that flexibility architecturally from day one later, it's so hard to change it and it becomes very rigid.
And then typically people end up rewriting the systems.
Richard Walker: 25:02
Yeah. Are you seeing a challenge with user adoption? Do people not believe this or not like it for some reason because it's not what they're used to. I. Change is hard, right?
Rabih Ramadi: 25:15
So far, not really. So far, not really. Because basically we understand exactly well what each persona wants in the system and what we build in terms of the workflow. The AI agents are very much aligned to this. We're not forcing people to basically change everything they do, but our premise is rather than doing it in an analog way, just doing pure digital, and now you have an AI assistant that will just make your life much easier.
Like rather than you calling or emailing to trigger an account opening, just talk to the AI system. Even eventually you can add a voice like, hey, I have lunch with my client. Based on that, talk to my assistant, say like trigger and distribution event for the account XYZ. And by the way, add an account with Pershing for XYZ basically beneficiaries. So we're just making life easier.
We have a lot of testing with the clients in terms of the behavior to make sure like per persona, we are really addressing what is important for them. And also like in the industry, people are very shy of exposing things to the end clients so they can participate in things like onboarding. And that's changing. And I would say that's the piece that has typically the most resistance in the past, which is the client doesn't want to do anything. They want to just email me the information and just do it.
It's changing now because now, especially with the new generation, people don't want to email you their banking statements and their Social Security number. They want to have a digital experience to do it. So we're seeing way less resistance to the change on that part than where it was three years ago.
Richard Walker: 26:57
Yeah, I was telling you before we started this recording that I've been working on software on the side because I wanted to learn how you could develop software with AI. It's just fascinating to me. I haven't done software programming in over five years, probably 8 or 10 years. And it feels to me the way I've architected this and I'm doing it, it feels like I'm just telling a robot, go do this. And it's a very intelligent robot.
It's often a toddler. I have to tell it to get back on the rails. But I like this feeling of I talk to my computer literally through my microphone, which types out the words because I'm using Whisper Flow, which is an amazing tool. And then it goes out and commands, and this agent goes off and does this work. And now I want it for Word and Excel.
And I just noticed, I think Claude just came out with they'll build spreadsheets for you in Excel natively. And I saw him sitting in the future. I'm seeing this like a point where we all just talk to our computers all day. Hey, go into my CRM, update this deal, we just got a signature, and changed the deal status. I don't want to go find all that.
It'll just do it right. So are you building it 100%?
Rabih Ramadi: 28:03
This is in probably the next 12 months. Rich. We think especially on the front line, people are going to interact with the app mainly from the chat experience. People don't want to learn. We think we have an amazing intuitive UI experience with different modules, whatever.
But we think in 12 months people don't even need that. People want to literally chat with apps. As this is becoming more and more mainstream, like the one you basically like as a cloud to create the Excel spreadsheet, this is what people are going to expect from apps. So I. Think so. We we this is exactly where everything that basically can be performed currently on the application in the next few months is all going to be done through that AI assistant interface, where literally you're typing or you're chatting to the to your assistant and doing everything for you, whether tell me everything I need to do before I climb eating or tell me, or trigger workflows for me, or give me some insights on XYZ segments of the market. All of this stuff is done now from the AI assistant, and that's how we see the future.
Richard Walker: 29:12
Now that's awesome. Okay, I want to ask a different type of question between you and your co-founder. Do either of you or both of you have deep technical skills in software development?
Rabih Ramadi: 29:23
Yeah, I am an engineer. I spent six years in engineering schools. My co-founder is even spending way more. My co-founder has a PhD at MIT in AI, so we both come from a very heavy engineering background, but we also understand the industry extremely well. So our thing is like we have a very deep technical background, but we know the industry.
would know what people look for. We know the problem of the industry. So we're very effective in using our tech background and applying the tech to specifically solving a business problem.
Richard Walker: 29:58
Do you think I really want to ask what it takes to start your business, or a business that says, I have a business problem and I want to use AI to build it from the ground up? Do you think somebody who doesn't have at least a software development background? I mean, you have a deeper connection with your partner. Do you think having a non-technical background, but having a good, solid business, understanding that person can start doing what you're doing? Or do they have to have a technical co-founder too?
Rabih Ramadi: 30:24
I think you need to know technical, because I think it's one thing to know what is the intended behavior of the software, but the foundation and how you do it makes all the difference. I can go and say, look, I want an AI bot that does x, y, z for me. That's one thing. And I can hire some really smart engineers, but you have to have at the founder level, a technical vision on how you're going to build something that is going to be differentiated in the market. Otherwise, you're going to build some features that quite honestly, will be very easily replicable because cloud in the future can actually, if I come up with something now that can populate Excel spreadsheets.
The example you gave me and how the cloud can do it. So you end up basically building features versus a foundationally foundational platform that will change the way people do things in life.
Richard Walker: 31:16
Yeah, 100% agree with you on that. So then do you think engineers are going to go away? Do you think you can get by without an engineer?
Rabih Ramadi: 31:24
I think the engineering mindset, and maybe I'm biased because I spent six years doing engineering studies. I think the engineering mindset will never go away. Yeah, I think AI is going to do a lot of what coders do. I think you probably won't need as many coders in the future because code generation, like our coders, uses AI tools like cursor, and they are three times more productive. But the engineering mindset that I think is invaluable for me. Like there's a certain way from an engineering perspective, you have to think about a process, about anything that I think will never go away from that perspective.
Richard Walker: 32:02
You know, I'm feeling that a lot of engineers are afraid to learn AI. And yet I don't want to lose my engineers. I just want them to get better at using AI to go faster, do more. I don't know if I want to hire more engineers, but I don't want to lose a single engineer. And if we could do three times faster, ten times faster, 20 times faster, then we could build that many more features.
I mean, imagine the ability. You may already have this ability, Rabih, but for me, it's a dream. A customer gets on a call and says, I wish I could quit doing this. And tomorrow it does it. Like when it was just me programming back in 2002.
I would do that over a weekend. Oh, let's build form groups. And then I'd show my partners, look at what I built over the weekend. But today with entire engineering teams. It doesn't work like that.
You have to go planning into a sprint and you have to do a release. I don't know, so I see it as super valuable.
Rabih Ramadi: 32:53
The engineering industry will be changing drastically in terms of and the best engineers already embrace AI and everything they do. Even our interview process. If you have engineers interviewing with us that are not using AI, that's always a bad indication because why not? Like the engineer is the one that leverages everything that currently exists and be 100 times more productive by leveraging what currently exists versus reinventing the wheel?
Richard Walker: 33:20
Oh, I don't want to hire a single person. I don't care if it's engineers or others who don't use AI.
Rabih Ramadi: 33:25
Exactly, exactly.
Richard Walker: 33:27
If you haven't used ChatGPT to solve a problem before, what are you doing? Like, I don't care. It could be Claude or whatever, but.
Rabih Ramadi: 33:33
Exactly.
Richard Walker: 33:33
How can you not function? I mean.
Rabih Ramadi: 33:36
Exactly. Exactly, exactly.
Richard Walker: 33:38
I don't know if I'm bleeding edge because I use it, I don't know, 20 times a day. Maybe I'm on the extreme, but I just can't see the world without it at this point. In fact, I had a nightmare a couple of months ago that all the llms went away and we had to go back to not having them.
Rabih Ramadi: 33:55
That would be a very hard adjustment. But yeah, like, it's so ingrained now in everything we do. And I have to tell you, Rich, it's great timing for us as a startup to leverage AI not only for this but for other processes in the back office like it drastically make a company way more productive. And actually, it helps you start the business much quicker than it was before.
Richard Walker: 34:18
Yeah. I was saying to my wife. Imagine if I was starting quickly today how fast we could go. And then I'm like, yeah, but we have revenue and stability and clients and brand names. So it's great.
It's a great time to be using AI.
Rabih Ramadi: 34:28
Yeah, exactly. Exactly, exactly.
Richard Walker: 34:31
Look, I have to get to my last question and wrap this up with you. But before I do that, what is the best way for people to find and connect with you and your company?
Rabih Ramadi: 34:39
Yeah, we have a website. It's still not the most elaborate website, but on purpose. We built a website. It's good enough for now. So that's one way we have a LinkedIn page also.
Events. Yeah, always happy to talk to partners, potential clients and how we can together improve the industry.
Richard Walker: 35:01
Yeah, you guys should take a look. I've seen a little bit behind the hood with these guys and it's. Man, I'm blown away. All right, here comes my last question. One of my favorite questions.
Who has had the biggest impact on your leadership style and how you approach your role today?
Rabih Ramadi: 35:16
Good question. A number of people had a big impact on me, but I would single one person. Her name is Debbie Leone. She was my first client at Goldman Sachs when I was in KPMG, and she was my sponsor and my mentor and my coach for 13 years, and she told me a lot of what I am now in terms of multitasking, in terms of doing the right thing for people, doing the right things for clients And and and also being in the details while being strategic, which I strongly believe in. You have to really know the details of what's running your business, while at the same time you have to have a long-term view in terms of how to build a business.
Yeah, I.
Richard Walker: 35:58
So appreciate that. I actually think one of my gifts is the ability to see the top level and the most detailed level at the same time, and look at problems and solutions and products, etc. from that perspective of the full spectrum. However, I struggle as the CEO sometimes. Do I get into these details or not? How deep should I be at this point?
Always. And I keep coming back. Yeah, I keep coming back to do it serve my customer to know this stuff or to be part of it? And if it does, I'm going to do it. I don't care what people think I should or shouldn't.
People might say, Rich, why are you building software on the side with AI? You're a CEO. I'm still a technologist. I'm an innovator, so why not?
Rabih Ramadi: 36:39
I totally agree with you. It's. That's the spirit of everything we do in Avantos. Like we're all hands on. Myself, my co-founder.
All. Everybody in the team. Nothing is too small. Nothing is too detailed. We're all in the details, which is what we think makes it a different company, which is what you guys do.
Also quick.
Richard Walker: 36:56
Yeah. Well, look, I want to give a big thank you to Rabih Ramadi, co-founder and CEO, co-CEO of Avantos, for being on this episode of The Customer Wins. Go check out his website at Avantos.ai, and don't forget to check out Quik! at quickforms.com where we make processing forms easier. I hope you enjoyed this discussion. We'll click the like button, share this with someone, and subscribe to our channel for future episodes of The Customer Wins.
Rabih, thank you so much for joining me today.
Rabih Ramadi: 37:23
Thank you.
Outro: 37:25
Thanks for listening to The Customer Wins podcast. We'll see you again next time and be sure to click subscribe to get future episodes.