What we talked about
In this insightful conversation, Hussain Ali Mirza shares his journey as a global product and operations leader with over two decades of experience across multiple countries. Currently leading Product Data & Customer Operations at Videomagic, Hussain explains how they’re using AI to transform long-form text into engaging short-form videos for productivity use cases. He discusses the challenges of working with rapidly evolving AI technologies, the importance of having humans in the loop when using generative AI, and how cultural sensitivity plays a crucial role in both international business and AI prompt engineering. Drawing from his experience working with teams across the US, India, Turkey, and Europe, Hussain offers valuable advice on product development, emphasizing the necessity of continuous customer feedback and understanding the existing processes before attempting to improve them.
Show notes
Welcome to Prevetted Podcast, a blend of tech and leadership! Today on the show, we’re joined by Hussain Ali Mirza, a global product and operations leader with over two decades of experience transforming customer experiences and scaling AI-driven solutions. Hussain currently leads Product Data & Customer Operations at Videomagic, where he’s leveraging artificial intelligence to turn everyday content into rich video storytelling. From managing global teams across the US, India, Italy, and Turkey to launching conversational AI tools and patented video compression technologies, Hussain brings a wealth of insight into building high-performing products and partnerships. Get ready for a deep dive into innovation, leadership, and the future of AI in customer engagement.
Full transcript
Federico Ramallo (00:01.974) Welcome to Prevetted Podcast, a blend of tech and leadership. Today on the show, we’re joined by Hussein Ali Mirza, a global product and operation leader with over two decades of experience, transforming customer experiences and scaling AI driven solutions. Hello, Hussein, how are you doing today?
Hussain Ali Mirza (00:24.92) Hi, Frederico. I’m good. How are you doing?
Federico Ramallo (00:28.246) I am happy to have you here. So yeah, yeah, absolutely. We’re very close. We were talking about that before the cameras start rolling because today I’m in the Bay Area and you’re very close to me. So that’s nice. Yeah.
Hussain Ali Mirza (00:30.042) Thank you very much for having me.
Hussain Ali Mirza (00:43.098) We should have this in person sometime.
Federico Ramallo (00:47.468) Yes, yes, I thought about it. I thought about it. It’s it will be a very Much better, but richer experience. So yes. Yes awesome, so let’s I have a few questions for you about your career. let’s start with you know, tell us a little bit about Your experience, you know in tech in india that’s where I believe you started right and then
you work across several countries, how did those international experiences shape your leadership style?
Hussain Ali Mirza (01:24.634) Excellent. So in terms of my experience, so straight out of engineering school, got into a tech space where we had built solutions on the web, which was in early stages at that point of time. Came to the US after that for a masters, which was in industrial management focused areas. And I focused my overall career in
delivering of technology, services, and products during the course after that. My experience in terms of geographies, both in terms of where I have worked and where I have delivered solutions has been across Asia, Europe, and the US. And obviously, the continents have
significant variance in cultures and how you would work in each of those, obviously. So the exposure to working and kind of delivering products to various geographies makes you very aware of cultural differences, of the priorities of cultures in each of these areas.
And for you to be able to succeed in those places, you need to adapt yourself to those geographies and cultures. Now, obviously, technology is a melting pot, especially around the COVID time period, where you are compelled to work remotely with various areas. So the head start of my career and my past has helped me a lot during that period of time.
Federico Ramallo (03:18.614) Interesting, interesting. I’ve been working with American companies, American teams, and teams in different countries in Mexico. And I experienced the same thing about, you know, found, you know, like you’re on both sides, you see the culture on one side, the culture on the other side, you can see the differences, right? And it’s interesting to be able to explain, you know, to both sides, like,
you know, to give more visibility and, and, you know, allow people to connect better because you can, you can help articulate both sides, right? And seeing that magic happening, it’s for me, that’s amazing. Right.
Hussain Ali Mirza (03:54.147) Right.
Hussain Ali Mirza (03:58.722) Absolutely.
Hussain Ali Mirza (04:03.298) It is. my everyday on my team is in the US, India, Turkey, and in some cases, Europe. So in addition to the challenge of finding a time for our everyday calls, which works for everybody, I think the…
A synchrony of all perspectives which come onto that call, both in terms of defining and executing on a product, makes it just a better experience with everybody’s viewpoint from the perspective they come on, which is influenced by their culture and their prior experiences as well.
Federico Ramallo (04:52.152) Yes. Yes.
So can you tell us a little bit what initiated your interest in product and customer operations?
Hussain Ali Mirza (05:05.178) Sure. So my early part of my career was in execution, project management, and those kind of short-term execution-oriented roles. And then from there, I had transitioned into operations, given the fact that once you are in a project role, you
get to know across the board all the areas which impact a specific project. And you come with an execution mindset of delivery and making sure all of your stakeholders are aligned and are happy with the outcome, which had taken me to an operational role from there. And once you’re in operations and I was in a customer service or areas, you
interact with your customers on an everyday basis, you get first-hand view of what they’re happy with, what they have issues or challenges with, and what they want to see out of your offering. You are the voice of the customer, right? So over a period of time, when you hear enough and experience enough, you automatically say that
if I had this, this, and this, it would serve our customer a whole lot better. So that had driven me to a product space where I took on the responsibility to build and deliver to what inputs I was receiving and the opportunities which I had heard. So even though I was in an outsource delivery unit, the value which we provided was
feedback to our customers, which is the enterprise business which we had serviced saying, your customers want this. This is what I’m seeing from interactions and data which we are working with every day. And this is what you need to do to get a richer product experience out for your customer. So at some point of time, I took it upon myself to try to
Hussain Ali Mirza (07:29.806) do that as a next step forward for me in my career.
Federico Ramallo (07:35.118) interesting.
I think that the understanding the customer, that’s the core of product. It’s critical, right? Because if the product team doesn’t understand what the customer needs, the engineer can build something amazing that nobody’s going to use, right? So the impact.
Hussain Ali Mirza (07:46.138) Thank
Hussain Ali Mirza (08:02.266) Absolutely. And it can be no-money, right? It can be, as you said, it can be what can make the customer’s life better or what you see is an obvious gap which your product is able to fill. Which the customer may not necessarily be aware of is something which they can use, but just by your experience in interaction,
you say that this can be a lot easier, a lot more efficient if we had this.
Federico Ramallo (08:38.222) I’m wearing multiple hats in my company. So one of the hats that I’m wearing is product owner. So I’m learning how to become a better product owner. But I’m coming from the software engineering side. So the part that where I’m struggling the most is understanding the customers because that’s more of a human skill.
Hussain Ali Mirza (08:53.22) Mm-hmm.
Federico Ramallo (09:08.107) in that process of developing that skill.
Hussain Ali Mirza (09:11.066) Thank
Federico Ramallo (09:12.078) So it’s very interesting. Or at least it’s a completely different skill set than building software, because when you’re building software, it’s you and the computer. So it’s different.
Hussain Ali Mirza (09:25.826) I come from a customer interaction implementation execution role. So I come from a frontline view, which has transitioned me onto a product role based on that interaction experience. I think that frontline view is very, very key in the ability to envision and to build
an effective product is that whole interaction with a customer because at feedback loop is extremely important.
Federico Ramallo (10:04.194) Right, yes, I agree. So, changing a little bit the topic, I want to come back a little bit on what you’re currently building. You’re currently working at VideoMagic. You’re using AI to turn plain text into video. Can you walk us through how that works and what the biggest challenges are?
Hussain Ali Mirza (10:32.92) Absolutely. at VideoMagic, we focus on is to create, one second there, I’m seeing an error message pop up.
Federico Ramallo (10:46.082) Yeah, yeah, take care of it.
Federico Ramallo (10:51.052) We are debugging life. No, I’m just joking.
Hussain Ali Mirza (10:53.262) Yes, we are. So at VideoMagic, we use AI to build a short form video for productivity use cases where you have documents, you’ve got a lot of long form of the text content, which is, we all know that for people to read through a lot of text is
a tedious effort. So we are focused on improving productivity as a use case, where we convert long form text to short form video, which can be used for marketing, training, awareness, and all those kinds of use cases. It works great for kind of summarization. And the most important aspects of this is to
while you are making it more consumable is not lose the essence of what a document or the original content is trying to convey. So where I sit in this is to extract, enhance, aggregate rated content which the engine needs to create that output video which is our delivery to our customer space.
So effectively bringing all the components together for it to be stitched together and delivered to the customer is where I come into the ecosystem from.
Federico Ramallo (12:38.882) Right, right, very interesting.
What do you think has been the biggest challenges of building this platform?
Hussain Ali Mirza (12:51.702) AI is an evolving space as we all know. It’s evolving extremely rapidly, but evolving nevertheless. What was there, say, weeks back could have changed dramatically with a new update at this point of time. It’s obviously getting better, it’s obviously improving, but the assumptions which you had started out with changed on a constant basis.
Most people use LLMs for their AI specific areas, but how you use it, the workflow which you build around it, and the effectiveness with which your delivery comes about, which is a part of the overall workflow chain, needs to be up to date with what’s available out there. And obviously the tools you used.
last month could very well be different from the tools you have chosen this month because of this is a rapidly evolving space and people are learning, catching up and fine tuning at the same time. And of course there are competitors which are looking for similar use cases which you are trying to do. So you always need to be a few steps ahead.
Federico Ramallo (14:14.424) Right. One of the things that I was reading about LLMs is the lack of uncertainty, right? Because when you’re a deterministic process, you have deterministic outputs, right? But here, when you have an LLM process, you really don’t know what you’re going to get, right? So, you know…
There’s always a risk of plus the hallucinations, right? But there’s always a risk of what you’re getting is not conveying what the user is looking for,
Hussain Ali Mirza (14:54.362) Right, 100 % agree and.
the human needs to be in the loop before things go out. And in an automation space which we are in, it’s hard to have the human in the loop at as many places as you’d want to have. So we make sure that through automation, we do as much of checking as we can do for
relevancy, appropriateness, et cetera, for content, because it’s also about interpretation, right? How you would interpret specific content would be how need not be the same on how an LLM would as well. So in our workflow, while we try to make it as easy as simple as we can for our customer,
We always recommend a customer have a look, verify before a final approval and obviously change as they need to be. They need to make changes and customization as and when they want it. So that built into the product as compared to leave it to a decision of the LLM.
Federico Ramallo (16:06.979) Right.
Federico Ramallo (16:16.034) Yes, yes, the more specific your request, the more predictable the outcome, but there’s always an uncertainty of unpredictability. And that happens to all AI power products, right? So yeah, and as you said, because all of this is evolving, right? I’m sure that this is going to be solved sooner than later, right?
Hussain Ali Mirza (16:29.658) Absolutely.
Federico Ramallo (16:45.716) yeah.
Hussain Ali Mirza (16:45.794) Absolutely, and the effort or the errors which we had a few months back is very, different from the errors we have now, right? And obviously it’s improved extensively. The tools involved has improved significantly. So that’s the part of the early question. This is a rapidly evolving space, predominantly positive steps forward to make things easier.
Federico Ramallo (17:16.366) So how do you see the role of generative AI evolving in the space of video generation?
Hussain Ali Mirza (17:30.522) It has a significant impact. was at a trade show recently, which is the National Association of Broadcasters, and the number of products, including by startups, in a generation space is significant, right from real-time subtitles, translation.
creation of a storyboard, creation of visuals based on those storyboards is significant. And the focus on these areas, because it’s…
It’s a part of the creative process which until recently had been people driven, as well as optimization in terms of cost and the amount of time which it goes in. So it may not be perfect, but it is definitely close to being kind of good enough, which is why you see a lot of adoption or at least interest in it as it is today. So we are in a generative AI space.
as well from ourselves. And the application is the product which we have in the market at this point of time. So it’s impacting everything. And it’s especially impacting the creative process. There’ll always be some level. There’ll always be some processes which need that fine-grained creativity and…
fine-grained enabling of a person, but there’s enough of low-hanging fruit which is available for people to use to optimize their work process. will help speeden up, it will make things in the proverbial better, faster, cheaper.
Federico Ramallo (19:36.514) Right. Yeah. I was on the AI marketing summit a month ago or so, and they were talking about how now they can use AI tools to edit video instead of having to go through the traditional channels, right? And they can, you know, so then what they were explaining is that we still are the creative, we still are
the person that leads the creativity. Now we have a tool that is naive creative, if you like, allows you to… You can tell this creative what you want and then it’s going to give you different options, but it is up to us to choose which one would make more sense to pursue and develop.
Hussain Ali Mirza (20:17.091) Absurd.
Federico Ramallo (20:34.626) So we’re still in charge of the decision making. sorry, Gaur.
Hussain Ali Mirza (20:34.97) Absolutely.
Hussain Ali Mirza (20:39.8) No, I’m sorry. had broken up in between, so I thought you had finished your thought. I think just to add to what you were saying, there’ll always, in terms of marketing or content creation for marketing, there’ll be that one perfect commercial which had taken hours to create and where you’ll spend most of your ad spend on to put it on all the platforms. But you also need a lot
of shorter content, quicker content, which you’re able to put out there to engage your user base as well. And that you can use AI to very easily create it to enable that interaction all the way through while you showcase what you have created by hand.
Federico Ramallo (21:30.178) Yes, I don’t think that the high production and high cost production is going to disappear. Yeah, because right now the level of detail that you can do with, generate with AI is limited, compared to the fine control that you have when you’re doing something live. But for…
Hussain Ali Mirza (21:54.989) Absolutely.
Federico Ramallo (21:58.036) a low budget approach if you want to do something that is good enough instead of having to spend 500,000 dollars you can spend just a few dollars and get an AI generated image that is going to be good enough for a much smaller cost if you’re willing to make those compromises you’re going to get a really good quality result that otherwise was
not an option.
Hussain Ali Mirza (22:30.106) Absolutely, and just to add to that, there’s a level of expertise as well, right? So, example, a small business will never be able, which wouldn’t have been overwhelmed by the amount of time and cost it takes to create.
an asset for them to use to market. Right now, there’s enough tools available out there where they can go in and they can create something of themselves. So it brings a lot more people into the ecosystem of what needs to happen, which wasn’t there earlier. So in addition to optimization, those of what is already in play,
they’re enabling a lot more other people to, a lot of other people to do things which they were not able to do earlier. Like even programming right now, there are various tools out there to help people write code. They may not know what to do with it, but there’s other tools as well to enable them to do those kind of things. I was seeing some articles somewhere where it says a lot of startup founders now who are not programmers themselves are able to use one of
coding tools to create an MVP or the basic POC for them, which they had to hire a developer earlier. So it’s bringing a lot more people who weren’t able to do things, now are able to do things if not as effectively as a programmer would.
Federico Ramallo (24:02.85) I am I have a friend that decided to refactor his code with with cloud 4 right It was magic, you know He had all these you know Object compacted, know long long long Pieces of code right he was able to refactor all that distributed on many objects, you know
It’s through the complexity and abstract in, you know, and seeing that happening was magic, right? Of course the code didn’t work, you know, but, you know, but it was beautiful to see, you
Hussain Ali Mirza (24:38.564) Hahaha!
Hussain Ali Mirza (24:43.258) Absolutely, and especially when you are trying to integrate with a third party platform, say your product, you can take API documents which is available of that third party platform, it in the system, get the code which you need to write and that helps you in your integration efforts significantly as compared to have to go through yourself and build custom code to integrate with platforms for which there’s good documentation out there.
Federico Ramallo (25:11.598) Yes, yes. every time I talk about this, that, oh, the LLMs are not as good as a senior developer, which is mostly true, right? And people talk about the quality of the code. But again, I come back to this idea of we’re still the drivers. We need to tell the code exactly what to do. And the more specific we can be, the better, right? So that’s
Hussain Ali Mirza (25:38.797) Absolutely.
Federico Ramallo (25:41.068) Daralla was, I mean, it’s mostly a reminder to myself that I say, you know, like, I need to be more specific, you know, I need to.
Hussain Ali Mirza (25:48.878) I had to press that’s a real like to your earlier question like at the start of this conversation about cultural interactions and sensitivities and working across various geographies. It goes back right to that point where you need to be specific on what you want in order to get what you’re expecting. And a good example which I keep kind of
of laughing about myself is when I first went to Italy and I went into a coffee shop over there. I asked for a latte because that’s what you do when you go to Starbucks over here. And this was a small place in a neighborhood. So the guy looked at me very, very intently and said, latte? So I said, see. So then he again asked me the same question. Then he had shrugged his shirt. He had shrugged.
his shoulders, he went, he poured some milk, heated it up and gave it to me. And I’m looking into that and like, like I’m expecting a cafe latte, right? And then in hindsight, it turned out that I need to be specific that I want a cafe latte compared to a latte which is milk. So, so that’s my one of those examples where unless you are specific, especially in a cultural
Federico Ramallo (27:02.168) Right.
Hussain Ali Mirza (27:11.354) kind of a scenario. You go to any Starbucks over here and ask for a latte, you will get exactly what you want, right? Which is not the case in my experience in Italy at that point of time. So it falls back to treat this as another culture, treat it as another language interaction where you need to be clear, you need to look at all possible interpretations in order to get what you want from the LLM or the system which you’re looking for.
Federico Ramallo (27:40.504) Yes, yes. And I mean, I know they’re getting better and better, but it’s still us that we need to tell the specificity and the clarity we need to provide that and the guidance. And then the other, again, I haven’t been playing with this, so I don’t know if it’s true or not. Well, a little bit. I played a little bit. But if you threaten the agent, the model, you get better results.
Hussain Ali Mirza (27:53.688) Absolutely.
Hussain Ali Mirza (28:10.03) did not know that and I should start doing that but because I have been cursing and threatening in my mind but not explicitly telling the model that but I should try that.
Federico Ramallo (28:15.758) You
Federico Ramallo (28:23.82) Yeah, and the thing is nobody understands quite why that happens, right? I mean, we build this amazing and complex technology that simulates how our process thinks, how our brain thinks, right? So, you know, that means that it kind of behaves like our brains, but we don’t understand quite why, you know.
Hussain Ali Mirza (28:53.838) That’s a good one, I should actually try it out. And I think just elaborate a little bit more on the previous point, right? I think when you are working with a chat GPT or anything else, just using a UI interface for yourself.
Federico Ramallo (28:55.298) Yes.
Federico Ramallo (28:59.758) You
Hussain Ali Mirza (29:06.394) you have the ability to iterate, where you said I was not as specific, et cetera. You can always go back and elaborate on it. But when you are using it as a part of your automation or your workflow, then you don’t have that ability very often, right? So you need to be as specific as you can and envision all possible scenarios. I’m not sure how I can include a threat in that statement, but I will try. That’s a good one.
Federico Ramallo (29:26.286) You
Federico Ramallo (29:32.736) Yeah, yeah, eventually becomes this, know, commented code, you know, do not modify because it works. We don’t know why, but it works, you know.
Hussain Ali Mirza (29:43.058) But then if they have an update after that and then it behaves differently under threat where it comes back saying like how say a call center when you call in and you make a statement like if you do that again I’m going to hang up on you and then they ban you saying Hussein has come back for a request he had threatened me last time let’s play around back with him.
Federico Ramallo (29:46.496) You
Federico Ramallo (29:58.456) Hahaha!
Federico Ramallo (30:10.06) Yes, yes, the I mean if you tell him some tell the the model something like do this like your life depends on it, right? Then you know, usually you get better results and the other one The other one that I learned is if you say take a deep breath And give me this response, you know again better results Nobody knows exactly why But it but it happens
Hussain Ali Mirza (30:40.28) I should try that. I have been going a courteous route so far saying and being as specific and eliminating all negative use cases to try to get my right answer. I should go down the path where take a deep breath, no work others for me. I’m in no hurry. I just want it right.
Federico Ramallo (31:00.75) I’m eager to see what you learn off that. And then the other one is if you try to give them, because I try that of giving them negative examples, don’t do this, don’t do this, because you see results and you say, don’t do this, right? But the way that the models are trained are on positive things. So now if you give them a negative,
Hussain Ali Mirza (31:05.978) I’ll let you.
Federico Ramallo (31:30.048) It takes them a while to feel, you you’re adding more complexity to what they need to feel out, right? The model needs to feel out.
Hussain Ali Mirza (31:38.808) I tend to do both, give a positive example and try to guide a thinking, but I also tend to put negatives in the areas I know it has failed on me in the past, just to try to bring it back on track. But yeah, I get your point.
Federico Ramallo (31:58.508) Yes. Yes. And I think that’s where the value of the product that you’re building fits because you can build a better rail, car rails around it. So you can have better results rather than just a generic LLM where you can type something. You might get a, it might generate a video, but it’s not going to have the quality that your product is going to have because
You have all that engineering around it to make sure that the output is better.
Hussain Ali Mirza (32:33.306) And also especially on specific use cases and the specific environments or the types of outputs which we are looking at. We aren’t trying to cater to the world, just trying to cater to the use cases which we’ve identified. So that makes it a little bit more guided as compared to anything else.
Federico Ramallo (32:54.69) Yes, yes, yes. Yeah, that’s the value that I’m talking about of everything you put around it. And you could be using the same model as a generic model, but with all that infrastructure around it, then the outcome is going to be way better.
Hussain Ali Mirza (33:15.894) And that’s the part which we had also spoken about saying do as much as in your point, guardrails, cross-verification, and as much as we can do through automation to validate a product before it goes in front of the customer.
Federico Ramallo (33:32.461) Right.
Federico Ramallo (33:35.822) What is a misconception, a common misconception that people have about using AI in generating video?
Hussain Ali Mirza (33:50.879) Hussain Ali Mirza (33:54.264) I think expectations. You have a variety of use cases where you
where the expectation, sorry, you have the expectation of a customer that they will get exactly what they want, right? And they feel that it is simple to guard to get the output to what they would think is very obvious. And this goes back to the guardrails and the training of the LLM and the…
channeling them to specific use case, where unless those are truly in place, you are not going to get what you expect. And also in terms of…
the output and what you’re able to do at least as of today, you still need a Tom Cruise to make a Mission Impossible movie to come out with everything which you can do, right? He’s still the one who’s hanging from that plane as an example. Go to that point where you can create that out of AI, most probably one day, but it will take a lot of human effort to guide the LLM
to create that exact video which will meet expectations of some people at least. It will not be a one-shot saying replicate this and give me the exact same output. There’s an effort which goes around it. The output may be extremely simple, but what goes into it to make that output happen is very complex.
Federico Ramallo (35:47.128) Right. Adding to that, people talk about prompt engineering. And now I’m hearing the term prompt alchemy. Because we really don’t know how that works yet.
Hussain Ali Mirza (36:05.594) Along with the AIS search and the LLMs, terms which are going along with it is also evolving over time. yeah, you need an LLM to figure out the true meaning of the intention of each of these as well.
Federico Ramallo (36:20.078) Right, right. Yes. So, so yeah, I, I completely understand what you’re saying. Yeah. So going back to your role, your role, if you could give one piece of advice to someone stepping into head of product, you know, for the first time, what, what, you know, what advice would you give them?
Hussain Ali Mirza (36:51.384) I think when you are in a product role,
The ability and the infrastructure to interact with your customers is very, important. You can’t build a product in a silo. You can’t, you need that.
reinforcement mechanism at all points of time to validate any assumptions which you are building into the system.
And it obviously depends on the product. If you are looking at automation or creating efficiency in an existing process, you need to have a good understanding of the existing process. What are you trying to replace? Because the existing processes, good or bad, have evolved over time for a specific reason. So you need to understand it in order to modify it.
And this is only one example, but from a product perspective, it’s very important to understand your target audience, your use case, and to build towards that as compared to do it in Asylum.
Federico Ramallo (38:14.348) Right. So team, you know, building your team, it’s key rather than trying to be the hero, you know, the solo hero, right?
Hussain Ali Mirza (38:30.042) I don’t think in general there’s a solo hero for anything. think any outcome, output, etc. has a lot of teamwork around it. Inbuilt into it, which is what makes the output happen. Though a product person can take credit for it in the end, it’ll be foolish to think he did it alone, right?
Just added to that, think stakeholder involvement, stakeholder commitment, a good feedback mechanism from the customer in addition, who is a critical stakeholder is very important because then you could be very surprised at the end of the product that this is not what helps your customer.
Federico Ramallo (39:26.702) Right. I’m coming from the software engineering side. So when I’m working with the team, our interactions is mostly software engineering with the product owner, Trying to get the requirements, right? So the way that I explain to my team, because as a software engineer, you don’t see the customer, right? mean, because your interaction is with the product owner, right?
So I tried to explain the software engineering team is what you’re doing, what you’re building is for the user. The product owner is, know, transmitting what, you know, what the user wants, needs, right? To the best of their abilities, right? So understanding that gives a purpose to what we’re doing, right? So we’re adding value to the users, right?
So what we need is to understand the user through the product owner. And that way we can ask questions. We’re talking about a third person that is not in the conversation, but we have to bring it in consciously. So that way, because I used to take the word from the product owner as face value rather than asking myself, is this a good representation of the customer?
or there’s something that the product owner missed, right? And how can we as software engineers add more value to it, right? To the process, right? Understanding the, you know, when we build a feature, why are we building that feature for? What is the value for the user to use that feature, right? And if we don’t understand it, then we can go and ask the product owner, right? And, you know, if the product owner doesn’t understand it, then we can, you know, rethink about, okay, are we…
Are we building something that makes sense or is something that we’re building just because? Right. Yeah.
Hussain Ali Mirza (41:27.802) 100 % agree. I think just to add to it, think depending on the product, obviously, every person involved in the product can and ideally should have a user perspective as well. And the hope and trust is that the product manager, product manager, product owner, etcetera has done a relatively comprehensive job to capture
all that a software engineer needs where most of their questions are already answered as inputs as well. But there’s always going to be gaps. This is a learning process. Most of the time it’s agile. You are learning on the go. So you are putting together a parachute while you, especially in a startup environment, you are putting together a parachute just after you have jumped off the plane. So…
Federico Ramallo (42:25.761) I’m
Hussain Ali Mirza (42:26.806) So everybody ideally is doing their best to capture everything which they need and should be doing. But there will be gaps which an error is best fixed as early as possible in the process than as a redo later.
Hussain Ali Mirza (42:49.304) or an oversight.
Federico Ramallo (42:52.696) So one question I wanted to ask you about, I know we’re going back and forth, so I apologize, but anyway, about video magic, are your plans for the future? What can you share? What are you planning to do in the next 12 months?
Hussain Ali Mirza (43:14.99) So a large part of what we do in VideoMagic is content creation out of other content which exists. So we are trying to create easily consumable content from long form content for wider distribution and an easy assimilation of a product.
So that’s our current focus is on productivity videos, be it explanation of a process, explanation of a concept, explanation about a product, and those kind of things from an enterprise viewpoint. And there’s why, and we all know that.
Video is the best form of consumption for anyone, which is why everybody’s always on their phones, looking at it as compared to holding it to their ear. So it goes to the extent and the power of video as a content consumption platform. So we envision the ability to take various inputs in whatever format.
and make it into an easily consumable short video as a use case for the near future. The use cases we cover will expand over time. The inputs which we take and what we do with it will expand over time. But we are committed to video as a platform and easy
easily understandable short videos as our output. Use cases will obviously expand. While we are looking at the enterprise productivity use case right now as our immediate focus, we have other use cases in the pipeline already which expands from that as a primary use case.
Federico Ramallo (45:31.212) Right, very interesting. Can you tell the audience what is the main use case that you’re focusing right now so that people that would fit that use case can go and review your product?
Hussain Ali Mirza (45:45.826) Absolutely, always a good thing to have. What we talk about a productivity as a use case is you’ve got documents, you’ve got blog posts, you’ve got a press release, you’ve got a news article which you may have, which currently is in text and long text, which you have out there but is not very well used.
People don’t have the patience, wherewithal, etc. in order to find it and go and read through it. We have the ability to take all such content, be it from a web page, be it from Word, PowerPoint, Google Docs, wherever it may be, and convert it into a short form video based either 30 seconds, one minute, or whatever, which you feel is…
easily consumed by your audience so that they watch a short video, they find interest in it and for further details will go back to a document to get the full picture. This will be the video will be used to whet your appetite and to give you a great summary which
is a precursor to you doing your actual reading after that.
Hussain Ali Mirza (47:23.898) So imagine I have a brochure I want to give or a white paper which I have to circulate. I might give you a one minute video to watch first, which makes you understand that you are interested in this. I will go for any further details to a document. Or you’ve got, say, a video or a document and say the employee handbook of
how do you do a performance appraisal for your employees, as an example, right? So if I give you a short video of 30 seconds with all of the snapshots, you have a brief understanding, then you can go back to the employee handbook and read through further details, look for the form, fill out the individual sections, et cetera, to move forward.
Federico Ramallo (48:18.05) Right, interesting, but you get the basic idea with the video and then you can begin. Yeah, or if it’s a potential customer, they can watch the video and they can say, that talks to me, that resonates with me and then, yeah.
Hussain Ali Mirza (48:32.698) Absolutely, this is exactly what I want. Let me get more details by contacting sales as an example to go from.
Federico Ramallo (48:38.872) Great.
Interesting, interesting. Yeah, so I will test out the product that you know how it works Great
Hussain Ali Mirza (48:46.936) I look forward for your feedback. Any feedback is good feedback. As I said, any customer input, be it a prospect or anything else, is a learning experience for the…
Federico Ramallo (49:01.666) Great. So I appreciate very much, Hussein, to be here. I think we learned a lot of new things, and I’m looking forward to see what you’re going to build next.
Hussain Ali Mirza (49:07.64) I enjoyed the competition. Thank you very much.
Hussain Ali Mirza (49:17.124) Me too. Thank you very much for your time. It was a great conversation with you and your audience.
Federico Ramallo (49:22.338) Awesome, thank you.