Brijraj Bhuptani on early US growth and lessons from focusing on outcomes first at Spry

What we look at is two three parameters before sprying. So the number of your revenue center is the number of therapists that work in the clinic. Overheads is all the other stuff and the other IT spend that you do. What we look at is the average revenue per appointment and the average revenue per therapist and the average revenue per employee. All of these factors we look at before sprying was implemented and after spying was implemented. We've seen clinics increase their bottom line because of colonists by a good 15 to 20 percent within few months of working with spying because a reimbursement sublitter because you build out everything the factor of indagent. Because there is so much automation you don't need the same number of therapists in the front of the staff who actually treat the same volume of patient. Hi, wherever you're joining us, I hope you're doing well. I come to tech conversations where we bring you insights from tech entrepreneurs, CXOs and investors. I'm Harir Klee and in this episode, bridge-rudge Bhuptani founder and CEO at Spray Therapeutics talks about how he and his co-founder Riyaz Rahman found an opportunity in helping small and mid-sized physical therapy clinics in the US to become more efficient and innovative with an AI-led revenue cycle management platform. Bhuptani also talks about pivots along the way and thinking about outcomes first that sharpened their go-to market focus and plans for going after bigger customers now all while delivering the product development and engineering completely from India. Okay, so bridge, thank you so much for doing this. Appreciate your time. For a sort of a serial entrepreneur as they like to say, it started in Jagan. So let's start with just a simple background about Spray Health and we'll go from there. Sure. My name is Harir. Again, thanks again for talking to me. I'm bridge co-founder and CEO of Spray. I'll start with Spray because that's my life at this point of time. It's one of the top rated software platforms for physical therapy centers in the US. From the outside in perspective of sitting from here, we think what's so complicated with the physical therapy center, right? The patient comes, gets physical therapy, pays money and leaves and that should be simple. But that's not the case more often than not, right? When the patient comes in for the first time, you have to get the paper work done. What is the injury? What happened? How it happened? What the goals are. Because we are focused on the US market, insurance, reimbursement for physical therapy, outpatient treatment. So you have to collect all the insurance and payment details because someone has to pay for it, right? And insurance more often than not pays for these treatments. You have to get the insurance details from the patient. Then the patient obviously meets with the therapist, does an initial evaluation, which needs to be documented and then sent to the insurance for reimbursement and also needs to be sent to orthopedic or a referring physician or someone like that for approval. That this is what the problem is and this is a treatment plan. If then sometimes the assistant could be doing the treatment, the assistant would write the document and the senior therapist will have to sign it up, right? So there's a lot more which goes in treating a patient, then the patient is walking in and going out and imagine doing this and physical therapy is a high volume profession. You typically for a business to make sense, every therapist will have to treat seven to eight patients a day, right? And the appointment ranging from 45 to 60 minutes. So you have to treat a patient and do all this administrative work and that's what makes it very difficult tiring for a therapist, right? So what Sprite does is using a combination of workflow automation, generational AI and integrations, third party integrations, which streamline the entire operations. So before Sprite, for example, if you have to check for insurance verification, you have to call the insurance company waiting line for 30 minutes and then get the answer. Sprite enables that digitally. You have to get papers, people to fill out the digital intake and then integrate into the main documentation. Sprite does that digitally. You send out the intake and it automatically integrates. It makes it easy for the therapist to write notes. So rather than you have to writing your notes separately, you just talk to the patient and the notes gets converted. And then more importantly, what Sprite does is make sure that all of that dovetails into your revenue cycle, which is the billing, the process of sending a claim to the insurance company and getting the money back. So think of it. We, the bigger goal what we do is let therapists be therapist, take care of all this administrative overheads burdens that they have. We拭line it for them and minimize the overheads for them. That's what Sprite does today. We've been around for three and a half years, only operating in the US, working with more than 150 clinics. I love today. Can you talk a bit about why you specifically chose physical therapy clinics? Sure. I think a journey for any entrepreneur is not straightforward. And I'm a second amount of entrepreneurs. The second I want to know, I think if I have to think about my journey the first time in the second time, the first one was high in emotion and low on business sense. I chose a statement I was in Bombay. I used to live in Bombay where traffic is a problem. Public transport is a reason. So I want to do something around public transport, road traffic management. So traveling Bombay would become easier. That's on grade. It's a great problem to solve. And I did beat it, solved it. But it's a difficult proposition to make money off because roads don't belong to you. Money of these government contracts. It transports the normal social service. So it doesn't make for good business sense. But it was a great business. We got great visibility. Great NPS course, etc. A lot of users are now. This time when I was starting a company, the second time around I was obviously very, very certain that I want to build a big business which makes commercial sense as well. Wanted to go after global opportunities. I had worked in Ola Pratt, between Riddler and Spray and at the opportunity to work in London, launch Ola in London. And I saw the difference in international market makes. I was thinking global day one. Thinking of was very sure that I want to be a big market and a growing market. So when I looked around different pockets of opportunities, healthcare is obviously the biggest healthcare in US specifically is one of the biggest price for any software engineers, software, but it's entrepreneur, right? And it's 4.4 trillion dollars. Just write them. I think even bigger than the GDP of India at this stage of time. So one of the biggest opportunities. But in healthcare is not one monolith, right? You have to think of healthcare as 1,000, 4 billion dollars opportunities rather than 1, 4 trillion dollars opportunity, right? So within those 4 billion dollars opportunities, I found physicalty have physical therapy. Why one of the biggest muscular skeletal physical pain, one of the biggest in the industry, like 25% of the US healthcare spend is in muscular skeletal problems, which is body aches, knee replacements, etc. And also growing. One in every two or three people have this problem. They need physical therapy treatment. Also if you look at the data, people are living longer, much longer than what they used to, but not necessarily healthy. So what this is, lifespan is increased, health is not increased, none of them need muscular skeletal care. Even our lifestyle just end, even the younger ones, right? We have said entry lifestyles, our food habits are different. So even they have started experiencing pains much earlier than what our predecessors did. So when I looked at it, huge market within that, we found a niche which was big, growing much faster. And we clearly saw technology can make the lives of physical therapist better, right? With automation, etc. And I personally am a big fan of preventive care. I personally am a fan of anything that does not need too much medication. I went through a personal journey myself where I lost 15 KT between Ola and Spray, right? And that essentially came through just good discipline, exercise, controlling nutrition, etc. So if that becomes a way of life for a lot, then physical therapy is only bound to grow. So a combination of, so the second startup, if I say, is more high on business logic and less on emotion. So the business, you've created that this will be a big market, growing market. And so far we've been proven right about it. When you started out, what was your idea for this product? And can you walk us through a little bit of what that product has evolved into today? Yeah, it's a very, very good question. And people ask me, right? What? Like, you are running, you are running a one of India's best consumer apps. And now you're running a back office operation for USL. How did that transition happen? Yeah, again, I think being a second-time entrepreneur, being wise and following with the opportunities rather than being fixated about what you want to do. We started a very different company. As I said, the idea was conceived during COVID. And I think the trigger, one of the triggers was my mom needing knee replacement at that time. And we could not get her replacement done because no one was operating. And also no therapists were willing to come home. And that's when the idea struck that why couldn't she do it? And she was doing it with my sister on Zoom, where it was a very, very painful experience for my sister as well as for my mom. And that's when I thought that why don't we democratize access to physical therapy? We had recorded content. We can, a person can stand in her laptop, follow the recorded videos. And using the camera, we could measure the movement and generate a report and send to the therapist. That's how we started. So we built that over a few weeks. And then when we took it out the market, the feedback was very different than what I anticipated. The feedback was that all this is great and futuristic, but we have a lot of urine now problems. So again, I think sitting from here, you don't understand the complexity of running a physical therapy center. It seemed very simple from here. But then you understand the complexity. Then you understand that the current solution that they provide is also very fragmented and broken. So before sprang, people using six or different pieces of software passed together. And somehow that didn't make sense to me. Like such a simple workflow of physical therapy, why should there be one all in composing software which should do this? Then I saw that the therapist was worried about the revenue. Like when will I get the money from insurance? I don't have any visibility. That's very counter to how a doctor operates in India, where you get the money over the counter and you're done, at least in outpatient services, even in patient for that matter. So all of that didn't make sense to me. And when I look around, I saw that there's a big, big void, that there is no solution which is solving from the doctor's business perspective. So a lot of the products which are developed in this industry are developed by the physical therapist or pseudo technologist, right? But no one was thinking business, that revenue is my outcome and I have to feel backwards towards it. So that's when I thought, that's where Spray came in, we started thinking, building, we started thinking revenue. I'm not saying you compromise on the medical thing, but built everything backwards from it. And that's how Spray became one of the first few companies to digitize the whole revenue cycle management because purely, which was mainly manual until then. And now a lot of companies are coming up, but that's how we started. So we thought business that, if I'm treating a patient, this is how much I should be able to make money, money. This is how my business should be. This is my cash flow should be. And this is how I get visibility. But for making money, I need to do these things. I need to build a EMR, which gives me a documentation. Then I need to build a schedule, which is scheduled for the patient. I need to do a eligibility, which will give me how much money to collect upfront. So we thought of money, build everything backwards, and that's how the Spray current evolution involved. And the first set of customers were exactly those, small and medium clinics. They were happy to pay us 5% of the collections, but we managing the entire workflow for them, right? So they can focus on therapy and we can focus on everything else for them. That's how the evolution happened, right? So it was more a case of not being fixated about what you want to do, listening to the market demand and actually catering to that. So we pivoted on the fly and the other thing I got from the patients is when you used to talk to the patients, you used to ask the patients, right? It wouldn't you like love it if you didn't have to go to the clinics and you could do it at home and the therapist would just give you a report. He said, no, we want to go to the clinics. We are like 65, if you look at a typical physical therapy audience, they're mostly 60 and above, 60 and above. They're mostly home, they're empty nesters, they don't know anyone. They want to go out, they want to go to the therapist, they want to talk. So the relationship is way beyond just getting some exercises done, right? And if that's the trend, then might as well catch on to that, rather than trying to change the world for no reason at all. That's how the, so we started with a very different product but evolved into what we are today after listening to the feedback and looking at ground realities. >> Three and a half years into the business, would you say that you are well-passed the product market fit? >> I would like to say so. Yeah, I think it's product market fit is something which is, it keeps on evolving, right? So do you, will I say that we have product market fit for the small and medium businesses, physical therapy businesses, which were early customers or early target, quite definitely is, right? I think we have one of the top rated physical therapy apps according to G2 and many of the platforms. Our attention numbers are, referral numbers indicate that and we are growing fairly fast, right? But when you're building a business in vertical, you have to also make a conscious move to continuously move up market because vertical by its nature is restricted market rate. You cannot be confined to a segment. So do we have the product for the mid market and the enterprises, yes, do we have the GTM no? So I just said the company keeps on evolving, the goal post also keeps on moving in the product market fit but from the early customers who are focused on billing, who for whom billing was important, do we have product market fit, definitely definitely yes. Do we have product market fit for the other cohort of customers which is lightly up market or for whom clinical data excellence is more important? Possibly no, but that's something we'll get to in the next few days for your month supporters, right? That's what I'm seeing. - How are you doing today as a business in terms of your ARR and all of that? - So I won't reveal that, but you're doing fairly. So you close to 200 clinics who signed up, right? And average ARR is anywhere from 15 to 25,000 depending on the size of the clinic. What we did in the first year of revenue, maybe we took, maybe I think two and a half, so three years to get to the first million. Now we are adding that in close to quarter. So that's what, so that's the scale we're operating at and we're growing fairly fast, right? We close to, you raised $20 million across, from investors from India to US, et cetera. Yeah, and that's the scale of the business. And if you look at G2 ranking, we are the number two and number three related physical therapy app software in the country within a span of two years. And I think we should be number one by the end of this year. - And you said that moving higher up in terms of your customer segment, going after larger customers involves clinical data excellence. - Yeah. - Can you explain that a bit? And maybe from there, we can also get into what kind of role you expect AI will play in that. - Exactly, great question though. So what happened, unfortunately, right? I think if you look at the history of electronic medical records and healthcare records, they were conceived to exchange information, right? About the patient care, interoperability between different groups, between different doctors, treating the same patient. Because your revenue is also linked with it, eventually we can be a billing platform, right? People started documenting because they wanted better reimbursements or better, we have better reimbursement so many shares coming. So the clinical part lost, it's the clinical documentation, right? Lots, it lost its purpose somewhere there, right? But when you go to the bigger groups, bigger groups don't operate in isolation. You're a bigger group who's contracted the hospital and your PT's, your physical therapists are working within the hospital along with other doctors. Now that is where you need to exchange notes, et cetera. You're getting referrals from NFL players, NBA players, right? And the coaches are interested in how your journey is progressing or not progressing. So you want to be able to document without the overheads, so obviously when you want to document, et cetera, it off, right? You want to document in detail, but you also don't want to take up a lot of time. And then when I'm reviewing cases, when I'm reviewing the patient journey, patient, I should be able to do those things fairly quickly, pictorially, et cetera, et cetera. I should be able to exchange data with other systems quickly, I should be able to marry the data with what other specialist is treating and make sense out of it. So making sense out of this clinical data, into making it interoperable, making it visually appealing, because remember, in physical therapy, a patient does not come for one-off treatment, like dental, right? In dental, you typically go for one, so you can read in detail, note it, figure out. Physical therapy spans over 10, 15, 20 appointments. That's when you get well. It's not like a one-shot treatment that you get done with. And you have to make sense of data all over the time and you have 20. So I think that's where we are spending time to answer your questions again. That's where Gen I comes in, right? A lot of data that is written by physical, every one physical therapist is different from another physical therapist. And not only physical therapist, every provider, every doctor, every provider, thinks of treating a patient, they don't have algorithmic way of treating a patient. They have a mental model, they have a mental workflow, which they execute, right? And they have a different style of documenting it. Some might write it in one way, the other way, the other way. So there is no standardization of that. So Gen I helps you standardize all of that, right? Because data is unstructured. Doctor one, mind, I did one way, doctor two. It's quite likely that the one same patient might be worked on by two doctors. So we're using Gen I to actually structure this data, provide in fashion which can be easily seen. So if an orthopedic is referring a patient to the physical therapist and the physical therapist is sending the note back, it doesn't have to be a verbose note, which you or she has to be. It can be some pictorial representative of what N20 visits, et cetera. You also normalize the template. So if two or three different locations are working on the same therapist, you can actually marry the data and start making sense out of it. So one of the powers of Gen I is that it can structure the unstructured, right? And a lot of data, especially in the clinical site, is in the unstructured domain. Is it unstructured by design or by the way it has been written? So that's what Gen I, one of the use cases, we are using multiple use cases of Gen I, this is one of the use cases of Gen I for us right now. Can you give me a few simple examples of the type of data that you use, the data on which you have trained the Gen I models for your purpose, what kind of data would it be? Would it involve some information about say the type of treatment somebody has gone through, or maybe the age group that they belong to, what kind of data would be important? Awesome. In your area of expertise. So I think Gen I, what Gen I does is we don't need to train the models anymore, right? I think the models are a large language models for a reason, they have a lot of training data, so we don't need to train them. What we need to do is do the right prompting, once we parse that data to them, which is unstructured data. And what we typically parse to them, we create prompts, basis on the workflow. For example, if it's a sports therapist, right? Sports therapist will talk about pain, range of motion, et cetera. Whereas someone who's treating headaches, versus tubular, we'll talk about dizziness, they'll talk about ability to sit and stand. So there are different workflows on how you evaluate a patient, right? You might just say everything, you might just speak out of paragraph, but what we use, we don't train the model, but we question the output of the model with the right prompt, so that the data can be structured in the format that the therapist expert or the hospital expert. So what we need for this is to have the workflows that the therapist are comfortable with. There are a lot of documented workflows. For example, in physical therapy, there would be something like the McKenzie workflow, but there are so many papers that I get published, there are so many literature, it depends on what school you go to, what professionally you work within your previous life. So everyone has a different workflow or a different format how they want to do that. So when a clinic gets onboarded, we ask them how they want to evaluate a patient. What are the step one, two, three, four, what are the four questions that add? What are the four assessments they do? Taking that data from the therapist, understanding that, that form and then using that to prompt that output of the LLM is what we do is how we actually make sense out of this data. Can you explain this some more to me? Who does the prompting and at what stage? From the time someone is looking up for a physical therapy center to actually go to on the internet, maybe they are searching. To the point where a sports maybe an athlete is actually meeting a therapist in one of the physical therapy centers that uses your product and so on. They sort of break it down to different places where these prompts are useful. Awesome. So I think the way to think about this, imagine a sports person, he or she has had an injury. First instinct is to go to an orthopedic surgeon. The orthopedic surgeon would do his own evaluation and tell him that okay this one does not need surgery or needs surgery. And then either if they don't need surgery they send the send effects to the physical therapist. This is my patient, ABC, this is age, this is the injury, why don't you treat him for 10 days and see where it's going. Or if the surgery has been done, they tell him this surgery has been done and this is the post-op that needs to be done. So it comes with the facts. So the first place where Genie is used and the prompting is used is actually analyze that facts. There are so many facts that are physical therapies get, because 70% of the patients are coming in through referrals from another profession. So the first thing we do is writing the prompts, first we analyze the facts type of facts. What type of, is it a referral facts? Is it a plan of care? There are different types of facts which keep coming. So the prompt is first to what type of document it is. Then depending on the type of document there are next set of prompts which say okay extract this information, extract the patient demographics, extract the patient injury, extract the patient insurance details. Those are all the details, extract from the facts and then insert it into the digital records as well. Then when the patient walks into the load and meets with the therapist, the patient and the doctor are having a conversation. The first evaluation is always how this injury happens. What was the moment pattern before what is it now? What are the verses? A free-willing conversation with happens between the therapist and the patient. And at that time they keep the recording on. So this entire conversation is being recorded. Then the therapist is asking the sports person to do some exercises, some moments to do an assessment of how bad or good the injuries. And when the assessment of the therapist is visually viewing the moment and then speaking into the mic okay the range of motion is 50, pain is 70, palpation is so and so. Which is again getting recorded. And then the therapist makes a plan of care that you have this and this injury. You will have to come to the clinics for the next three weeks, two times a week. These are the treatments that we will be doing. All of this is spoken like how we would typically speak to a patient. What surprise doing is recording all of this. Once the patient once the doctor says I'm done. We first of all segregate what was the patient what the patient was saying what the therapist was saying. Then we divide into three parts. What was the patient problem? What is the objective measurement? What are the measurements that the the pair doctor did? And the third is what is the assessment of the doctor view? So we structure all of this data first into patient information, objective information and the assessment. And then within that we find the next level of field. So within objective what was the pain scale? What is the range of motion? What is the palpation within assessment? We find out with the right prompts. How much what is the frequency of my coming? What are the modalities I'll be treating with what are the exercise I'll be doing it. So we have the first level prompt and then there are sub prompts within that to extract all that information. So that's how we create a note. And then obviously a lot of the prompts that are created are from the template of the workflow that was initially given to us by the doctor when we started onboarding them. And then when this note is completed, we put it into a data store which is easy to read, easy to consume every unit for insurance company and for the referring provider as well. In the absence of your solution or another solution like this, this all would have been manual. Yeah, very here to type out a lot. So people who have been working on spine on and so earlier it used to take them 15 to 20 minutes to write a note. Now it takes anywhere from 2 to 5 minutes. So you're saving around 15 minutes on average on a note. And you're as I said, you're treating around 7 patients a day. So you're easily saving around 90 to 100 minutes per day on just this simple administrative process or writing a lot of other administrative processes. But just the simple process of documenting is where you save a good 9200 minutes every day. And you give us a sense of a little bit of how you're measuring all the benefits that using this prior brings to your customers and how does that tie into your pricing models. If there's any evolution happening in that in general in the industry as well. That's a great question. What we look at is 2 3 parameters before surprise. So the number of your revenue center is the number of therapists that work in the clinic overheads is all the other stuff and the other ID spend that you do. What we look at is the average revenue per appointment. And the average revenue per therapist and the average revenue per employee all of these factors we look at before spring was implemented and after spring was implemented. And this is we've seen clinics increase their bottom line because of all of this by good 15 to 20% within few months of working with spring because a reimbursement subator because you build out everything the factor of integral. Because there is so much automation you don't need the same number of therapists and the front of staff to actually treat the same volume of patient. So reimbursement per appointment reimbursement and the average revenue per employee is how we measure the effect of spring and revenue and we put our money where amount is right. Our revenue model is continued on that. So we have for any clinic working with right there are two two things they pay for. One is they pay for the software the base subscription model conventional slash model but for the whole money that we get from insurance we take a percentage of the revenue. So the clinics make hundred dollars we make four to five dollars out of them. So we are also a partner in their journey. So we are our interests are aligned with them right it's not like a typical software company which you some software to them doesn't matter whether they can use it or not whether it works for them or not and you get the money on a recurring basis. So we are a partner in two cents and we are are interested aligned right in making sure that at least the revenues are maximized. So can we now sort of logically if this progresses to what you said earlier about going after larger customers. Yeah. Can you explain to me in the world of physical therapy centers what would the typical large customer look like and then maybe we can talk about going from where you are today to having many such large customers. What is the next big evolution needed both in terms of your product and your go to market strategy as you called it. So large is one I think fundamentally comes down to two factors how many therapists are working within your organization and then how many patients are you seeing within the yard right and there are multiple models in physical therapy there is one is to one model where you typically see one patient at a time and there are models where the therapist is actually spending them it to or three patients are the same time. So large clinics today are sweet spot is so we started off with clinics who are actually single people one to three PTs right today's spot is clinics who have five to 20 therapist because we are partners with them because we are sizeable chunk percentage of the revenue our account values are fairly high right so clinic even clinic making a million dollars for example. The revenue for us would be around $30,000 significant and there are lots of million dollars next today that's a sweet spot clinics making a million dollar revenue today that's where bulk of our current customers are making and they employ anywhere from five to 20 therapist. I think the next evolution for us is clinics with a 10 to 50 therapist right the 10 to 50 therapists slightly bigger centers either owned by a family by a franchise where slightly well their family by a franchise or by small P forms that's the next target for us. In terms of product do we have to do anything different at this point not really I think a products ready we in fact have a few middle level customers is just that we need to add add them to our kitty far more consistently so I think the better evolution is needed a GTM process where we figure out how to manage the stakeholders how to navigate that selling a bit market we sell a slightly longer sales cycles procurement involved IT security involved how to navigate those meses. So I think the 10 to 50 is where we do that and after that the 50 to find it will be the journey which feel hopefully ending in the next year but it will be more an evolution of mindset and more evolution of a sales process than anything obviously when we want to move beyond the find it which will also involve evolution in a business processes. This is we being able to a positioning as well and that's something we'll do sometimes in the course of this year but right now I think the immediate target for us is to evolve as a sales process which is also hiring the right sort of people who have done this in the past you can tell help us to these mid market slightly longer sales is where the biggest interventions and needed today is your sales already a combination of remote selling from India plus the Dond Street field deal and so on. No we don't have any field on street impact so it's a combination of the lead generation discovery calls happening from India and then the account executive is in US taking over and closing out. So it's it's a tag team between the India sales team and the US sales team and they work in harmony and they share the spoils of the deal together as well. So a lot of the and many of the small deals which are one two three clinics are close from India as well. We don't need anyone from the US to even step into those and actually close out the India public is in India itself. So bulk of the sales personnel and R and the logos are being closed from India but in terms of value obviously the account executive is in the US to every girl because the high value deals are closed in collaboration with them. Over all how many people on your team? Yeah we have 160 people right now. 80 of them are in the services part of the business which is the revenue cycle management billing etc and the remaining 8 year are typical SaaS companies, this product, sales, marketing etc. And you do like 100% product development engineering everything from from India is it? Yeah yeah and engineering is my standard I was a CTO full-up act of this and Bill Ridler is one of the best products I think in the last decade as well in a very difficult industry. So building complex products, building high-skill products is something I have tied right and also have the because my experience in Olaidran I have the network of bringing the right people to help me in that journey as well. Can you talk about that a little bit more building products? How has AI in our affected that or how is AI changing the whole process of building software products itself? It's a great question and I'm still forming an opinion on that I'm doing a lot of experiments in my view what it does is that what I mean AI has been around for a while I think you're referring to more genie right. Genie has democratic democratize access to data sciences, machine learning AI and the previous generation right. If I compare my experience with what I did in the last decade when I was building Ridler we spent up so much we spend so much time on the prerequisites like building the pipeline setting up the data pipeline, procuring the training data, training it, tree training it and by the time you get to the actual problem to solve you end up spending so much so much more time right and Genie just makes that easy right you don't need to worry about training there is so much the large language models have so much data and you can actually start working on the business problem statement from day one. So what has changed is a lot of your time to market or time to validation has drastically reduced when Genie is what I would think a lot of the redundant work lot of the boring work is what I would say right which did not give or I would say non-belible work because you had to do that to get to the final product but you could not essentially use that as a final product is taken away you so getting the first version of the product out getting the proof of concept out validating whether this will work not work how will it work has actually the time to that has been shortened drastically so that's what I think Genie does very well second what Genie does is I think it again access democratize access to knowledge right and you don't need those specialist people to tell you how if there are some people in a company who would know a certain way and a nuanced thing would go like how an insurance would work how to do that certainly Genie automates a lot of that right so you don't you know dependent on the special skill people who who would be sometimes ready to find or not easy to work with you can automate that I think what has not changed is clarity of business you still need the same clarity of business you still need to be able to imagine what the customer wants you still need to be able to understand what would make the customer happy what would give the customer is going so that is still the same but a lot of the work which was needed which would slow you down which would in the initial set of dependence on people has reduced and as so therefore making things to go like much faster I still have to make an opinion on whether we can use this yet technologies or the new generation AI to maintain things to you enhance things from the current state to the next state I don't know I have not done much work there but getting the first version of the first couple of versions out I think yeah it does great Genie I specifically does a great job with that would you put this next class of AI which you all are calling agentic AI or AI assistance that does much more than answering questions but actually tries to do some tasks for you would you consider that as part of the Gen AI sort of you know AI and within that your engineers are you all already you know you have begun to accept code that is suggested by these agents and incorporating it in your code with humans reviewing it how is that evolving so I think agentic AI is one of the manifestations of Genie AI and I'll just give an example I think 12 years back I had a job where I had to crawl websites of other e-commerce companies because I was working for any commerce company and do the price comparison for them right and you had to write different crawlers as we would say for each of those websites today you don't need to do that right then agent you can just extract in simple English that you go to website AI website V website V website CX I said do this and they would do the same thing without you having to incur that additional development for each of those modules etc that's why I said the time to make things happen is much much faster so that's I think the second question was that you were asking right that how we we started incorporating Genie AI so a lot of the work for example creating videos for tutorials which needed graphics which needed specialized video animation all that is now being at least tutorials which we give to our customers is being done by Genie AI lot of code reviews code conflicts which needed human intervention is now so we're using co-pilates within that so we're not using AI to write the code yet completely but we're using it to expedite a development making sure the accuracy and the cleanliness of code is much higher as we develop new products we evaluate using them from scratch but to be honest we haven't yeah we haven't had done that yet right but we hope to do that sometime do they say but right now we're using more it as a co-pilot then a creator is what I would say what I would say and can you circle back to when you started out as as a first-time founder if I can ask you to talk a little bit about your entrepreneurial Genie what were you doing at that time why did you decide you wanted to be an entrepreneur and not work in a company yeah so yeah it's a great good because I don't come from a conventional background of IT mechanism which we will find the first quote of startup entrepreneurs to be right now I can move any mediocre I can I graduated from Bombay University even to the US they can hear their student doing those time was working on H1B and I thought that was it right I think that was the peak it's just that when I was working I was fairly good at tech so I realized during my work and always felt and fulfilled as an employee so I worked with three companies before starting surprise I worked in Qualcomm as an embedded engineer I was not happy I think I did great there but I was not happy the pace it was just a big company so I joined the startup in the same space in San Diego itself and did reasonably well they built system of some scratch which gave me the confidence that I can build things around up I don't need an infrastructure doing be successful but that eventually shut down the startup didn't do well and then I came to India and started the capital centers for seers so I had this mix of experience across right a big company is start up and then coming to India and starting a US company in India is catacaptive it's scaled to 1200 people across all of this within the first year I was fairly bored right I thought I found things very easy for me and not mentally stimulating and there's always this is that I'm not being fully utilized is what I thought right and I could do a lot more what more I don't know but I should I am capable of doing more and I should be doing more is the inherent feeling I had with every job that I was the three years in the three jobs that I have to starting up that's how it triggered I think that's undercurrent was always there and when I paid off my home loans you were financially stable I don't know I think you were okay taken care of is what I would say then I started it take the plan and see why not right I think the idea again I think a big mindset difference when I I work with Laavish very closely and very ambitious right from day very clear on why they want to do things and that that changes your that shapes your thought process and shapes your action for me it was let's give it a try if you don't do it now it will never happen so it was very defensive it was good that I started up but it's a very defensive mindset right not the right mindset I should have started with the way they're okay I want to do this because I want to change the world I want to do something make something meaningful make something big obviously that's happening now when I'm doing right but that was not my mindset when I started that like I'm not happy with this let's try this and let's see where it goes it was very incremental so the out so the evolution of the company was also very incremental the eventual output was also good not great is what I would say but it all starts with that mindset so my mindset was not what it should have been for a young guy starting up at that time is how it was but frustration with corporate jobs across different countries that is hard to do more but I think that those are the trigger points for me to start up right at that time okay okay so let me bring you back to today at Spray over the next 12 months, 18 months, what would be your top priorities? Top priorities is to increase our GTM momentum. It's sales and marketing. I know you ask this question. We have well-passed the stage where we have product market fit for the current set of customers you want to target. We have good momentum on sales, how to build on that? The market. We are the top, if not the first, best physical therapy software in the top two or three physical therapy software. And at our pros point, the best value for money softwares in the industry. So how does all of that translate into a significant market share? Is what I'll be focused on? In parallel, as you say, as we all know, there is a platform shift also happening. Genie is there, it's for you. We are also evolving all our interfaces, all our platforms to be compatible for the next wave of user experience. I don't envision my kid or any of the younger guys coming and clicking on the sash tool to book an appointment or write a movie. It will all be either was searched, wasp, etc. So getting ready for that shift when as in when that happens happening very fast is what we'll be doing in the background. And that's something we'll see to you as we'll all be focused on. For me, the biggest priority will be how to do just tips to a product. And the opportunity we have and comment your rate with the market share that we think is possible fast. That's it for this conversation. I hope you found it interesting. You can find all our podcasts at fobesindier.com and on your favorite podcast app. I'm Hari Arkely. Thank you for listening.