Welcome to a new episode of LA PIPA Studios, where we explore the cutting edge of Data Science & AI through conversations with industry leaders. Our goal is to spread scientific knowledge and technical expertise while diving into real-world applications and insights from our guests. In this episode, we talk to Ganes Kesari, Co-founder and Chief Decision Scientist at Gramener. An entrepreneur, AI thought leader, author, and TEDx speaker, Ganes advises executives of large organizations on data-driven decision-making. His expertise lies in the application of data science to solve business challenges and in building teams that promote a culture of data. “I advise executives of enterprises, NGOs, and governments on organizational transformation using data science. I specialize in building data science teams and helping firms adopt a data culture. Starting my career in technology consulting, I spent a decade helping Fortune 50 clients use technology to enable and transform their businesses. My tryst with data happened about a decade ago when I co-founded Gramener. It has been an enriching journey at Gramener so far. I’ve experimented with the practical aspects of startup building and have been observing how our culture continues to evolve. Advising businesses on the application of data science, I’ve been deepening my skills in data analytics, machine learning, and information design. I’m fascinated by the big-picture view, complementing it with hands-on execution and an eye for detail. I’ve spent a lot of time strategizing and laying out the direction for Gramener—then leading from the front by executing that strategy, building teams, and delivering value to clients. I take exceptional interest in discovering deeper patterns, simplifying things, and presenting information visually. I challenge myself to explain AI and technology without math or jargon. I rejoice whenever I manage to design visual stories that distill complex insights onto a single sheet of paper.” Thank you for sharing your expertise, Ganes! Your perspective on building data science teams and fostering a data-driven culture provides invaluable guidance for organizations looking to harness the power of AI and analytics.
4o
4o
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:06):
Thank you for Independent Data Science Day specialized in data
driven business change. In this podcast, our guests great knowledge
and experience with our listeners.
Speaker 2 (00:30):
Good morning, Gonness. How are you doing today? Maybe good afternoon?
Speaker 3 (00:34):
That's right, so good afternoon, Swiss. So this is yeah
close to afternoon.
Speaker 2 (00:41):
Now awesome.
Speaker 4 (00:42):
How are you doing. I'm guessing I'm guessing that you're
a very busy guy. We'll get into why a little
bit later. How is your agenda for they looking like?
Speaker 2 (00:53):
What are they looking like?
Speaker 3 (00:54):
I think if you keep the schedule a part and
just enjoying the Midsummer long base. So so this is
part of the year where you have more hours in
the day to do stuff. So I'm enjoying that.
Speaker 4 (01:06):
Would you mind briefly introducing yourself, Knness, so that those
listeners our conversation get to know you not only related
to your business grummer, but more about your origins, where
you're from, educational.
Speaker 2 (01:21):
Background, how you started your.
Speaker 4 (01:23):
Career and then you became an entrepreneur, a little bit
of all of this.
Speaker 3 (01:29):
Sure, So I have about nineteen years of experience, all
of it in technology solving business problems. I started off
my career as a programmer coding in main frames Gobol
those father is, so over the years, I've been playing
(01:51):
several roles and been playing a lot of project program
management roles. About ten years back, me and my co
founders we decided to start something of our own, and
that's how Gramner was born. So at that point I
was leading technology implementations and managing program managing businesses for
(02:14):
verticals like healthcare and finance. So when the idea of
Gramner came about, we were looking at several challenges in
the industry. The one which we were very excited about
and which we saw resonating the market was a problem
of consumption of data. There was a lot of data
(02:35):
in the market and they were at that time design
talking about twenty ten eleven. Enterprise bi tools were available,
but they were not delivering the right actionable insights for
decision making. So that's the problem we wanted to solve
at Gramner. We got started and our value proposition was
(02:56):
with the data that you have, we can help you
identify the right insights, present them as stories, and enable
strategic and the right decisions across your business. The same
value proposition and the mission we are continuing with today.
While a lot has changed in data and analytics space.
(03:17):
The challenge is interestingly, the ability to make decisions and
the disconnect from data has not changed over the last
ten years. So that's a problem we see with clients today.
So that is something that we continue to help our
clients with today. And the role I currently play at
GRAMMA is that of leading advisory and innovation. So my
(03:41):
title officially is Chief Decisions Scientist. What it really means
is helping our clients make better decisions with data. In
our advisory practice, I help organizations identify how to layout
a data roadmap, how to execute it, and how to
get the best ROI from data. So that's what our
advice and implementation practices uppoth and innovation. We have AI
(04:04):
labs and story labs, experimentation teams. I lead this mandate
so that we continue to stay at the edge of
what is happening in the industry, do their experimentation and
apply the right relevant techniques to solve organizational challenges.
Speaker 4 (04:19):
First question, so, based on what you've just described, even
though your mission or your proposition value as a company
hasn't changed that much, the market and the maturity level
of the companies did change.
Speaker 2 (04:36):
I guess at ten years ago.
Speaker 4 (04:38):
The majority of the business we're still focused on building
their own data warehouses and putting a reporting layer on top.
A few years back, they started to, you know, launch
more advanced analytics iniciatives, trying to create analytical models to
you know, forecast and predict and act in a proactive
(04:58):
way rather than react if way. So how did your
go to market offering change over time? Because I'm putting
myself in yours us because I'm feeling in your suits
in the sense that two years before we started betrock,
and in two years many of the statements that we
(05:20):
usually use during our peach have changed. Sort of say,
just because technology changes very quickly and new compareditors emerge,
you see that use cases for different industries are.
Speaker 2 (05:37):
Easier to sell.
Speaker 4 (05:38):
Sort of say, so, how have you adapted the go
to market offering to what you need to do today?
Speaker 3 (05:47):
That's a good question. Over the last ten years, the
market has been writing on the data analytics and AI boom,
so this has been look back at. Our early years
started off by educating people on what is data visualization,
definition of data visualization. I remember the first couple of
years we did a lot of educational sessions as part
(06:09):
of our marketing. Whereas now, thanks to the coverage of
media and the general praise or hype around AI big data,
people understand there are investments already in this space. What
has not changed is being able to leverage these investments
(06:29):
for business decision making. There are a lot of tools today.
Earlier we had just Excel and say just to the
enterprise bo tools. Today we have data visualization, fancy visual
stashboards available. Earlier we had very basic descriptive analysis. Today
we have machine learning, predictive analytics. What has not changed, however,
is how do you use these new technologies, these fancy
(06:53):
tools to help your business user make a decision? That
the challenge we saw ten years back continues today, so
our go to market the pitch has changed in terms
of linking all of these ecosystem of tools. But ultimately
we are talking about the focal point of decision making
(07:14):
using data. We're really talking about how, but the investments
organizations have made in data and AI, how you can
leverage that and get ROI out of it. So it's
not about education of why you need to do it,
but how you can do it effectively with the investment
you've already made.
Speaker 2 (07:34):
Okay, so you've seefted in that way.
Speaker 4 (07:37):
But now looking at these services plus maybe products that
you've peeled over the years, what's the positioning of positioning
of grammarar right now, I've looked at the website.
Speaker 2 (07:50):
I've spoken before with you.
Speaker 4 (07:53):
But to whach extent you create a log analytical developments
and do each extend or which percentage of your revenue
comes from, let's say so as a service kind of
projects because you've ut staff in the past that you
can utilize again and adapted it to another environment. And
(08:16):
how much do you create from a scratch based on
a strategical development that the client needs because he is
quite a dog.
Speaker 3 (08:25):
If I talk about our value proposition and what we
offer in the market. So it is while I mentioned
that solving business challenges using data, so we bring in
there are three elements to it. One we have a
low code platform which is a rapid application builder. So
if you want to create a data science application for
(08:47):
improving customer experience or retaining employees, so whatever is the
business purpose, using a low code platform, you can create
the right data science application that brings in analytics, the
deep analytics capability, which is the insights part and visualization
information design which is a stories part, which is what
(09:09):
we call as insights as stories through a low code platform.
So that is our positioning and what we offer to
the market. Now in typical engagement organizations, when they approach
us with a business problem, most often we end up
creating a specific custom data science solution for them an
(09:31):
application using the Localde platform. So there is a consulting
element to it and there is a platform element to it.
So that's where you have the platform software as a service.
Subscription revenue is one part. The other one is the
services revenue which is coming in from consulting for solving
that particular problem, but using our own lo code.
Speaker 4 (09:52):
Platform and is your low code procession. And similar to
what I don't know other visers out there are doing.
I mean I can name a few H two data
I q mL studio.
Speaker 2 (10:07):
I guess that whenever you have.
Speaker 4 (10:09):
A conversation with lead, a potential client, or you're just
in a workshop educational workshop, you may be given these
or these names may come up. So what's the main
difference against them?
Speaker 3 (10:22):
But the key differentiator is a lot of these art
platforms which don't offer the the implementation application part of it,
so we bring in the platform and we bring in
the consulting. So that's one one difference from a pure
played platform provider and from a pure place service provider. Again,
(10:43):
that's a differentiation that we have a platform which is
which can stand up a business ready tool in a
matter of a few weeks as opposed to custom build
using which might take months. So to answer your question,
it is the advisory plus implementation combined with the low
(11:04):
code platform, which is which is where we actually not
just share the tool but also solve the problem. That
is a differentiation.
Speaker 4 (11:12):
And do you have for did you have any specific
specific focus in in terms of industries or even business
units you know, I don't know marketing, mixed modeling, for media,
marketing teams, any any any of these were more common
when you're thirty. Then this has changed over time. I'd
(11:33):
like to have a view on how you you evolved
in terms of these services and use case that you've.
Speaker 2 (11:38):
Built across the industries over the past ten years.
Speaker 3 (11:42):
We started domain agnostics. Over time we built a competency
based on the footprint and based on the demand in
the market. So we have four verticals today, which is
one is and one is ESG the nonprofits social governance.
I think that's a big one. The work you've done
in sustainability by your diversity smart cities, all of that
(12:02):
falls into ESG space. That's one vertical. And then we
have pharma it's another big vertical. We work with organizations
like New Artists and other large PARMA players. And then
we have supply chain and logistics, and fourth furness technology
and consulting technology organizations consulting firms like Deloyity and hy
(12:23):
So these are the four verticals we solve problems for
and within these verticals that could be a variety of problems.
For example, for one of the technology players, a large
technology laptop manufacturer, we have done multiple things, one for
their customer experience team, and we have done work for
(12:44):
their marketing team. So like there could be several horizontals
or several teams within an organization we work with, but
all of this falls into for example, farm or any
of these four verticals.
Speaker 4 (12:55):
Primarily, I'm curious because when I was reading or going
through your website and all of the content that is within.
Speaker 2 (13:02):
It, I did see that you refer to your services.
Speaker 4 (13:05):
As been design led, meaning I guess that you follow
a specific design methodology whenever you try to adapt that
low code solution to the business challenge.
Speaker 2 (13:18):
Right, I mean we Aspetrock.
Speaker 4 (13:20):
We also follow a few methodologies and we establish a
framework with a client where we try to, you know,
gather not only or set the expectations and gather insights
at the beginning, but we also exchange continuous feedback through
the development of the solution. Is this somehow related to
(13:40):
what you do when you're trying to implement that low
code solution based on what the business needs your client needs.
Speaker 3 (13:47):
Yeah? Absolutely yes. So what we mean by design letters
there is consumption at the root of everything that we do.
I mentioned that was the problem we set out to solve.
So consumption challenges solved by using story telling and information designed.
It's not important just to build a model which say,
can work on geospatial data and it runs AI algorithms,
(14:11):
but importantly you need to interpret it and for the
end user to recommended decision saying that this is what
you need to do and present it in an engaging
way using stories. So stories, when you say that is
a focal point, it cannot be an afterthought. You need
to have that as the front and center in terms
(14:32):
of where you start the discussion. So when we talk
to clients, we talk about what are the business priorities,
what decisions we need to enable and how we can
enable it for a particular user, and what are those
points that we need to how can we design that
So that becomes the starting point and the entire the
(14:53):
framework or the model we've built for delivery is inspired
based on that, and we have a a framework which
is inspired by Agile in terms of continuous cycles of development,
and we have over the last ten years we have
built our own methodology delivery methodology. Broadly, we call it
a radar framework, which is the overarching framework which covers
(15:19):
everything that we do at gram Now, one of the
key elements of that is, say, when it comes to implementation,
the design lead approach, storytelling lead approach. So there are
several toolkits to use under this broad methodology. That's how
we operated GRAMM.
Speaker 4 (15:35):
Very interesting, and I mean you've been in Grammar since
the ficking in because you're one of the founders, so
you most likely established the foundations of the initial delivery team.
You were mentioning that you use an Agile framework. But
since you adapt that low code solution to each one
(15:58):
of the cases that you find or that you encounter.
I have a few questions, do you I mean also
to compare your way of doing things against our way
of doing things and see what we can.
Speaker 2 (16:10):
Learn from each other.
Speaker 4 (16:11):
Do you have separate sprints for each client and account
because ourselves we do have a common sort of meeting
every week where we put together everything that is happening
across our portfolio. And then they have different squads for
each of the clients. Because we may puritude as fts,
(16:31):
some of our consultants data engineered data scientists that later
are part of the individual sprints that happen within the clients.
Speaker 2 (16:42):
What's your scenario like.
Speaker 3 (16:44):
As is inspired by a jet? We still have those
milestones which are closer to the traditional waterfall kind of
meteorology in terms of requirements design development. But in each
of these we have multiple multiple iterations that we run through.
For example, if when we're building this, there, when we're
building this. I talked about the initial design thinking workshop
(17:08):
where we try and sketch out what the solution should
look like or where should they even build, where should
they even apply data and analytics? So those are the
initial foundational discussions, and after that understand looking at the
data and understanding what is the right approach, how data
and analytics can be applied in this approach to solve
this business challenge. And then information design, how design can
(17:30):
come in, data visualization and machine learning algorithms can come in,
so all of those. If we talk about those initial milestones,
they we identify that and then they run iteratively. But
once we finalize that this is the approach and this
is the final deliverable likely to be, then that is
when we start multiple sprints in terms of running these
(17:53):
sprints to develop different versions of the application, so using
the lowcode platform, as there is a list of about
say sixty features to be implemented, so we have this continuously.
Every week, we have some features being developed. Initially it
could be a combination of some critical back end data
(18:14):
connection features, and then there would be a skeleton being
built a UI layer, and then there will be some
charts which will come in after that. So progressively you
see that being built and we have regular reviews of
this solution as it is being built with the client stakeholders.
So it's a combination of a child and some of
(18:34):
those standard milestones, just that we do it iteratively within
primarily when it comes to development and also in the
design phase there are iterations they run through.
Speaker 2 (18:44):
My question was only it related to how you deliver that.
Speaker 4 (18:49):
For a client, but also I mean you have different
projects that are being run in parallel and you have
a theme of how many data professional sort of close close?
Speaker 3 (19:02):
Two hundred and fifty is our team strengths?
Speaker 2 (19:04):
Wow? Congratulations?
Speaker 4 (19:06):
Well, first of all, congratulations, but if you have one
hundred something professionals there that are split across accounts, if
you say like that, do.
Speaker 2 (19:17):
They work not in insulation?
Speaker 4 (19:19):
But are they gathered by squads based on the challenge
the business challenges that they are facing and the solution
that they are building. And then do they have some
sort of unified or is there someone who has a
unified view. I guess there will be some similar role
to a CEO OH that has an overall view across
(19:42):
the whole portfolio projects.
Speaker 3 (19:45):
So, firstly, the unifying the process of the methodology is
there is a delivery framework which we follow religiously across
every single project and this is currently in the sixth version.
We launched the very first version. I think the second
year of operation. It's currently version six is going on.
So that is a mature framework depending on whether it
(20:09):
is a basic analytics like a diagnostic analytics project, or
a machine learning or a deep learning kind of a project.
Depending on the project, it is adaptable and we have
ways to execute the delivery framework, and we have account
leads people who oversee the relationship dot com and also
(20:33):
they have ownership for the solution the effectiveness of the solution.
So assuming there are five projects running for a client
and that person has that single point of view in
terms of how we are touching the different parts of
the business, how they all come in together, because often
there is there is synergy between these applications. So that's
how we have that ownership. And at the overall organization level,
(20:56):
when you talk about across all of these accounts, across
all of the client accounts we have been working with
about one hundred plus clients. Across all of these accounts,
we have a delivery head who brings in that the
complete organizational view to look at what is going well,
what is not going well, and how we need to
adapt our delivery framework to the new technologies which.
Speaker 2 (21:17):
Are coming in.
Speaker 3 (21:18):
So that's how we manage and keep it up to date.
Speaker 2 (21:22):
Okay, got it now? And based on your experience, I
mean now that you've hired.
Speaker 4 (21:27):
I don't know if you were involved throughout the hiring
process of all of those one hundred fifty something people,
but based on your experience, what makes great data scientists
a great data engineer? I know that being well at
being good at programming at Python or using I don't
know the tech stack of us here maybe useful. But
(21:49):
apart from that, what are the soft skills that you're
after whenever you are.
Speaker 2 (21:57):
Helping with the interview process.
Speaker 3 (21:59):
So there are five roles which we hire for and
have written about it in my articles. Broadly, it is
the data consultant role, functional data scientists, which is statistics,
machine learning, data engineer which is development operations DevOps and
programming role. And we have the design expertise information designers,
(22:21):
and we have project managers. So these are five key
roles we sta for and each of these, each of
this has a specific skill set and mixed coming into play.
I think based on the primary area I mentioned it
will be obvious. But to your question, what is that
most important skill we look for across all of these roles.
It is the application of technology to solve a business challenge.
(22:46):
So you might have someone who's great in deep learning,
but they are not able to understand the interpretation of
it to a business problem or how to apply and
solve solved for a specific scenario. Then that skill is
not put to the best choose. So we always look
for application skills, and that's a common challenge we've seen
in the industry. How you can bridge the technical skills
(23:08):
with what is needed on a project and can people
walk a few steps and apply their skill set. So
that's what differentiates from a great individual from someone who's
good enough or average in their skill set.
Speaker 4 (23:23):
I agree, But at the same time, what you're saying
is extremely hard and very unaffordable sometimes because you may
get someone coming out of university or has a few
years experience, maybe two or three, that is very good
at using technology and at understanding technology. But since hear
or see hasn't spent that much time facing business problems here,
(23:49):
see hasn't built that business document because it takes time,
doesn't it. So how do you get hold of this
talent before it gets too expensive.
Speaker 3 (24:01):
We have a mix of freshers, people from campus and
the latters. So when it comes to freshers, we invest
time in getting them on board and training them for
these skills so that we have good content we have
developed in the house, and we also put them through
very tailored shadowing projects and on the job training. So
(24:23):
we have figured out a way to use these people
and develop these applied skills for freshures. When it comes
to laterals, we're looking for that's something which we test
as part of the interview process to the best extension possible.
But that's something which we constantly if we see some shortcomings,
we again mentor and support people, but through a combination
(24:46):
of channels right when it comes to laterals, either through
assignments of practices or through scenario based questions. So that's
how we try and identify the fitment.
Speaker 2 (24:57):
I mean something that we do worse helps when we
try to do that. I mean, we we've.
Speaker 4 (25:03):
Built a close relationship here with the university and we
try to not much but link the academic world with
what's happening in business right because sometimes it's lacking behind.
I mean, the the programs around technology are not so
(25:26):
up to date, and we try to make sure that
you know everything that happens in university really allows the
students to hit the ground running whenever they join us.
So we created link between us and the university and
it really helps us to you know, build the specific
(25:47):
expertise in different technologies that that helps us.
Speaker 3 (25:52):
Yeah, that's a great point. In fact, I'm glad that
you're doing this. This is something that we have been
doing since that first year of inception as well. We
have partnered with several universities and universities across different disciplines,
whether it is the machine learning engineers, or data scientists
or even the design information designers. We have MOUs with
(26:14):
many universities and there are many where we have helped
them design the course, the curriculum and we run guest
lectures or in some cases we have run full like
full semester, handled the complete lecture. So we continue to
get involved. And this is across geographies in in India,
in the US as well, and we've seen that make
(26:35):
a big difference because often we go back and hire
from these campuses. We develop a relationship with these campuses
and hire students from there, so that also helps to
bridge some of these gaps even before a person comes
on board. That's a good point.
Speaker 4 (26:50):
Yeah, if I may, what's the split of stuff that
you may have based in the US or or even
around the world against India And where did the company begin?
Speaker 2 (27:02):
Did you start in India?
Speaker 3 (27:04):
Yeah, we started in India. For about five years we
were based exclusively in the Indian geography, and after that
we've branched out. Today we have presence in North America,
in the US, Canada, and Singapore, and so there are
a few other places where we're seeing greater interest for
our offerings. But if roughly, if we look at the split,
(27:27):
majority of the people, large majority of the people are
in India, I would say about roughly about ninety percent
of the teams in India and other global locations.
Speaker 4 (27:38):
And I guess that you open in your offices headquarters
for just locating people remotely in Canada, North America, and
all of that is because you find more business opportunities, right,
more commercial opportunities.
Speaker 2 (27:52):
There is a larger market to tackle.
Speaker 3 (27:55):
That's right, ya, closer to the client and yeah, h.
Speaker 2 (28:02):
Okay, I know.
Speaker 4 (28:06):
The next question that I'm going to ask would take
a couple of hours of debate, even though I'll try
to be very concise. If if if I were to,
I mean, as I said, we started better a couple
of years ago. Were started with a strong premise in mind,
(28:27):
which is, we want to make change and technology easier
for our clients by you know, using data science as
a tool that makes them operate better and more efficiently.
Because we felt that in the past, data science and
(28:47):
an AI was perceived as something very techy, very deep tech,
far from being business practical, and we wanted to change that.
So probably linking the world of the scientific technical experise
and what's providing ro O I to the business. And
I feel that that's something.
Speaker 2 (29:08):
That you've tried to do over the past ten years.
Speaker 4 (29:11):
So if you were to sure world learning with me
around your entrepreneurship.
Speaker 3 (29:19):
Past, which one would it be? I think that's a
good question. There's a lot which which can be added there.
But let me if you have to share just one point, right, Uh,
it's just uh, staying resilient and staying in the market.
So over the last ten years, we have seen a
lot of technology trends come and go, and the needs
(29:41):
of the clients also change. If you, if you stay
true to your mission and do not get swayed by
those short term either peaks or the trucks. For example,
in the early early years there were this there was
a strength around big data, and there was a lot
(30:04):
of investment going in that direction. And then suddenly there
was about five or five years back, AI enterprises started
investing a lot more. So many trends would come and go,
and you might see some short term business coming in
that direction. And when that trend it becomes a fat
and goes away, that business would no longer exist. So
(30:25):
there will be a lot of ups and downs. My
recommendation would be stay true to your mission and if
you have chosen a big enough problem, a beginnough vision,
that won't change despite whatever small the hypes come and go.
So stay true to that. And then if you're able
to stay in the market, because often in this space,
(30:50):
and that's the case with many of the businesses as well,
business to business enterprise selling, the cycle time is usually long,
and often you establish in relationship and eventually the project
will come in only six months later or a year later,
or in some cases it comes like three to four
years later. So we have seen cases where six years
(31:11):
after we made the first contact, a person reaches out
and says, hey, we were impressed by what you showed
six years back. Are you ready to do a project now?
So that staying in the market and staying proved to
your mission is what matters. That would be my recommendation,
Thank you very much.
Speaker 4 (31:27):
And based on that, I guess that something that you
said as a goal from the beginning is steering away
from only being perceived as a software service, plug and
play solution and being considered considered as a strategic partner
right in that data science sort of path at your
clients are brave or trying to get ready to take.
(31:51):
So how do you do this because it happens to
us right Sometimes it depends on the engagement and it
depends on the level of antuity of our clients. But
some of them they request a piece of development that
helps their data science.
Speaker 2 (32:08):
You need or it you need to move ahead when we.
Speaker 4 (32:12):
Really prefer, you know, as a service or professional services
company to build a long term relationships.
Speaker 2 (32:18):
So how hard was this for you?
Speaker 4 (32:21):
How did you ensure that you say you state in
that strategic path, the strategic positioning.
Speaker 3 (32:27):
That's a great question. Usually the projects might start transactionally
with with a small need or there is a burning
problem a client wants to solve there within the next quarter,
so that's how they reach out. And unless you are
a big established player in an industry, people may not
come to you for a long term relationship from the outside.
(32:47):
But what you'll have to use to your advantages when
these initial transactional or short term engagements come in. Take
it as a challenge to show enough value in that
so that you get into the radar of this client
and after that continuously keep up selling, move up and
talk see how you can MAC to the right stakeholders,
(33:10):
the right sponsors, and if you can demonstrate that you
can solve a bigger problem, show examples, show your past
case studies, and talk value proposition as opposed to solving
a transactional challenge. So if you're able to upsell and
demonstrate that big thinking and MAC to the senior stakeholders
and use that language, I think that steadily you will move up.
(33:30):
In some cases will be very rapid second third engage
you can get there. Others it might need some more time,
but you'll eventually get that that needs the focus that
you're very clear what problem you want to solve and
where you want to MAC. But you're patient enough to
work through the projects and work your way into that
high level mapping quick one.
Speaker 2 (33:51):
I know my answer, but I want to know yours.
Do you really date as high ends?
Speaker 4 (33:56):
Analytics are so powerful that it can transform many aspects
of our society. Society, I mean, I've seen that it can,
and I believe that it will continue to shape society
as a whole.
Speaker 2 (34:10):
I mean it's in everything.
Speaker 4 (34:11):
I think it's in autonomous driving, it's in our own devices,
it's in decision making. But to which extent do you
believe data science or advanced analytics If you want to
use the password AI because or it falls under the
umbrella somehow of data science you can too, to which
extent is going to change our world?
Speaker 3 (34:33):
I'm both excited and worried about the potential of data
analytics and broadly AI. It is very powerful. Having worked
in the space for over a decade, I see, I
see the potential, and at the same time, I'm worried
that it doesn't There aren't enough checks in the market,
and the incentives for taking it in the wrong direction
(34:56):
based on short term profits is too enticing for many
organizations and many individuals. So it is very powerful, but
it could post a big threat. So what I'm excited
about is there is a way to do this right
and this can be that invisible force which influences and
(35:19):
shapes everything that we do, whether it is our personal life,
how we lead our lives and how we make personal decisions,
or at an organizational level, how you run a company,
how you sell to clients, or even big ticket issues
like climate change and other environmental aspects. I think this
can play. This can play the role of a catalyst
(35:40):
and can be that invisible influencer to move things in
the right direction. But what we need to be clear
about is are we aligned on the right outcomes and
is it equitable fair? And we have some minimum regulatory
checks and self impost checks in place with that people
don't misuse it. So I think that I want to
(36:03):
see this evolved, and I want to see more checks
also come into place so that this can be sustainable
and it can really create a better future of everyone.
Speaker 2 (36:13):
Awesome.
Speaker 4 (36:13):
I'm glad to hear that you're taking those measurements sort
to say, I think they you In April published also
some guide guidelines, and I guess the US and different
parts of the world will continue to do so at
a government level. But I'm really glad that you bring
this up. Okay, yes, trying to wrap up. I think
(36:34):
it's been a very interesting chat with you. I really
enjoyed it also because I feel that I'm sure in
the same shues that you were in a few years ago. Again,
congratulations on what you've done with Graminar. I think building
a one hundred and fifty people business, it's something that
I do admire. Some last question, any recommendations on future guests.
(36:59):
I know that you that you've been part of dead
x talks, so probably you know a few interesting people
who should I v is picking with about the data
science and AI and the future of this practice.
Speaker 3 (37:11):
There are several interesting people, and that's one reason I'm
active on LinkedIn, because I see I bump into a
lot of great people, great content. I can definitely send
you a long list of people like I can shoot
that over to you or email. But there's a lot
of people doing good work. And I see also this
thought leadership thing taking taking shape where people are sharing
(37:35):
value and they are doing it selflessly. So there are
many interesting individuals to talk about. And in terms of
the earlier point you mentioned at Bedrock, I think I've
seen your work and you're moving in the right direction.
So very excited to see you grow. So good luck
in your journey and we're happy to have this conversation
(37:56):
with you.
Speaker 2 (37:57):
Awesome. Thank you so much. And last one read It
wasn't the last one. I need a book.
Speaker 4 (38:02):
I need something to read that you know inspires me.
Do you have any book in mind that you think
I should be reading in the next few weeks.
Speaker 3 (38:11):
I've been reading. The current book I'm reading is Competing
in the Age of Ai. Okay, professor his name is
Kareem Lacani. He's at Harvard Business School. That's an excellent
book which talks about building an organization in the current
age where you have algorithms pretty much taking over all
aspects of the business. So that's an interesting book. Gives
(38:32):
great examples of new age companies and how some traditional
and old companies are reinventing themselves. So that would be
my recommendation.
Speaker 2 (38:42):
Thank you very much. It's a great.
Speaker 4 (38:46):
I think it sounds really good. I need to get
that and I need to start looking at it. Okay,
thank you, Ganesh. I really enjoyed our call. I really
enjoyed recording this and hope the best for you and
for Grammar.
Speaker 3 (38:59):
Thank you, Thank you for Take It by Uri, play
like
Thank you for Independent Data Science Day specialized in data
driven business change. In this podcast, our guests great knowledge
and experience with our listeners.
Speaker 2 (00:30):
Good morning, Gonness. How are you doing today? Maybe good afternoon?
Speaker 3 (00:34):
That's right, so good afternoon, Swiss. So this is yeah
close to afternoon.
Speaker 2 (00:41):
Now awesome.
Speaker 4 (00:42):
How are you doing. I'm guessing I'm guessing that you're
a very busy guy. We'll get into why a little
bit later. How is your agenda for they looking like?
Speaker 2 (00:53):
What are they looking like?
Speaker 3 (00:54):
I think if you keep the schedule a part and
just enjoying the Midsummer long base. So so this is
part of the year where you have more hours in
the day to do stuff. So I'm enjoying that.
Speaker 4 (01:06):
Would you mind briefly introducing yourself, Knness, so that those
listeners our conversation get to know you not only related
to your business grummer, but more about your origins, where
you're from, educational.
Speaker 2 (01:21):
Background, how you started your.
Speaker 4 (01:23):
Career and then you became an entrepreneur, a little bit
of all of this.
Speaker 3 (01:29):
Sure, So I have about nineteen years of experience, all
of it in technology solving business problems. I started off
my career as a programmer coding in main frames Gobol
those father is, so over the years, I've been playing
(01:51):
several roles and been playing a lot of project program
management roles. About ten years back, me and my co
founders we decided to start something of our own, and
that's how Gramner was born. So at that point I
was leading technology implementations and managing program managing businesses for
(02:14):
verticals like healthcare and finance. So when the idea of
Gramner came about, we were looking at several challenges in
the industry. The one which we were very excited about
and which we saw resonating the market was a problem
of consumption of data. There was a lot of data
(02:35):
in the market and they were at that time design
talking about twenty ten eleven. Enterprise bi tools were available,
but they were not delivering the right actionable insights for
decision making. So that's the problem we wanted to solve
at Gramner. We got started and our value proposition was
(02:56):
with the data that you have, we can help you
identify the right insights, present them as stories, and enable
strategic and the right decisions across your business. The same
value proposition and the mission we are continuing with today.
While a lot has changed in data and analytics space.
(03:17):
The challenge is interestingly, the ability to make decisions and
the disconnect from data has not changed over the last
ten years. So that's a problem we see with clients today.
So that is something that we continue to help our
clients with today. And the role I currently play at
GRAMMA is that of leading advisory and innovation. So my
(03:41):
title officially is Chief Decisions Scientist. What it really means
is helping our clients make better decisions with data. In
our advisory practice, I help organizations identify how to layout
a data roadmap, how to execute it, and how to
get the best ROI from data. So that's what our
advice and implementation practices uppoth and innovation. We have AI
(04:04):
labs and story labs, experimentation teams. I lead this mandate
so that we continue to stay at the edge of
what is happening in the industry, do their experimentation and
apply the right relevant techniques to solve organizational challenges.
Speaker 4 (04:19):
First question, so, based on what you've just described, even
though your mission or your proposition value as a company
hasn't changed that much, the market and the maturity level
of the companies did change.
Speaker 2 (04:36):
I guess at ten years ago.
Speaker 4 (04:38):
The majority of the business we're still focused on building
their own data warehouses and putting a reporting layer on top.
A few years back, they started to, you know, launch
more advanced analytics iniciatives, trying to create analytical models to
you know, forecast and predict and act in a proactive
(04:58):
way rather than react if way. So how did your
go to market offering change over time? Because I'm putting
myself in yours us because I'm feeling in your suits
in the sense that two years before we started betrock,
and in two years many of the statements that we
(05:20):
usually use during our peach have changed. Sort of say,
just because technology changes very quickly and new compareditors emerge,
you see that use cases for different industries are.
Speaker 2 (05:37):
Easier to sell.
Speaker 4 (05:38):
Sort of say, so, how have you adapted the go
to market offering to what you need to do today?
Speaker 3 (05:47):
That's a good question. Over the last ten years, the
market has been writing on the data analytics and AI boom,
so this has been look back at. Our early years
started off by educating people on what is data visualization,
definition of data visualization. I remember the first couple of
years we did a lot of educational sessions as part
(06:09):
of our marketing. Whereas now, thanks to the coverage of
media and the general praise or hype around AI big data,
people understand there are investments already in this space. What
has not changed is being able to leverage these investments
(06:29):
for business decision making. There are a lot of tools today.
Earlier we had just Excel and say just to the
enterprise bo tools. Today we have data visualization, fancy visual
stashboards available. Earlier we had very basic descriptive analysis. Today
we have machine learning, predictive analytics. What has not changed, however,
is how do you use these new technologies, these fancy
(06:53):
tools to help your business user make a decision? That
the challenge we saw ten years back continues today, so
our go to market the pitch has changed in terms
of linking all of these ecosystem of tools. But ultimately
we are talking about the focal point of decision making
(07:14):
using data. We're really talking about how, but the investments
organizations have made in data and AI, how you can
leverage that and get ROI out of it. So it's
not about education of why you need to do it,
but how you can do it effectively with the investment
you've already made.
Speaker 2 (07:34):
Okay, so you've seefted in that way.
Speaker 4 (07:37):
But now looking at these services plus maybe products that
you've peeled over the years, what's the positioning of positioning
of grammarar right now, I've looked at the website.
Speaker 2 (07:50):
I've spoken before with you.
Speaker 4 (07:53):
But to whach extent you create a log analytical developments
and do each extend or which percentage of your revenue
comes from, let's say so as a service kind of
projects because you've ut staff in the past that you
can utilize again and adapted it to another environment. And
(08:16):
how much do you create from a scratch based on
a strategical development that the client needs because he is
quite a dog.
Speaker 3 (08:25):
If I talk about our value proposition and what we
offer in the market. So it is while I mentioned
that solving business challenges using data, so we bring in
there are three elements to it. One we have a
low code platform which is a rapid application builder. So
if you want to create a data science application for
(08:47):
improving customer experience or retaining employees, so whatever is the
business purpose, using a low code platform, you can create
the right data science application that brings in analytics, the
deep analytics capability, which is the insights part and visualization
information design which is a stories part, which is what
(09:09):
we call as insights as stories through a low code platform.
So that is our positioning and what we offer to
the market. Now in typical engagement organizations, when they approach
us with a business problem, most often we end up
creating a specific custom data science solution for them an
(09:31):
application using the Localde platform. So there is a consulting
element to it and there is a platform element to it.
So that's where you have the platform software as a service.
Subscription revenue is one part. The other one is the
services revenue which is coming in from consulting for solving
that particular problem, but using our own lo code.
Speaker 4 (09:52):
Platform and is your low code procession. And similar to
what I don't know other visers out there are doing.
I mean I can name a few H two data
I q mL studio.
Speaker 2 (10:07):
I guess that whenever you have.
Speaker 4 (10:09):
A conversation with lead, a potential client, or you're just
in a workshop educational workshop, you may be given these
or these names may come up. So what's the main
difference against them?
Speaker 3 (10:22):
But the key differentiator is a lot of these art
platforms which don't offer the the implementation application part of it,
so we bring in the platform and we bring in
the consulting. So that's one one difference from a pure
played platform provider and from a pure place service provider. Again,
(10:43):
that's a differentiation that we have a platform which is
which can stand up a business ready tool in a
matter of a few weeks as opposed to custom build
using which might take months. So to answer your question,
it is the advisory plus implementation combined with the low
(11:04):
code platform, which is which is where we actually not
just share the tool but also solve the problem. That
is a differentiation.
Speaker 4 (11:12):
And do you have for did you have any specific
specific focus in in terms of industries or even business
units you know, I don't know marketing, mixed modeling, for media,
marketing teams, any any any of these were more common
when you're thirty. Then this has changed over time. I'd
(11:33):
like to have a view on how you you evolved
in terms of these services and use case that you've.
Speaker 2 (11:38):
Built across the industries over the past ten years.
Speaker 3 (11:42):
We started domain agnostics. Over time we built a competency
based on the footprint and based on the demand in
the market. So we have four verticals today, which is
one is and one is ESG the nonprofits social governance.
I think that's a big one. The work you've done
in sustainability by your diversity smart cities, all of that
(12:02):
falls into ESG space. That's one vertical. And then we
have pharma it's another big vertical. We work with organizations
like New Artists and other large PARMA players. And then
we have supply chain and logistics, and fourth furness technology
and consulting technology organizations consulting firms like Deloyity and hy
(12:23):
So these are the four verticals we solve problems for
and within these verticals that could be a variety of problems.
For example, for one of the technology players, a large
technology laptop manufacturer, we have done multiple things, one for
their customer experience team, and we have done work for
(12:44):
their marketing team. So like there could be several horizontals
or several teams within an organization we work with, but
all of this falls into for example, farm or any
of these four verticals.
Speaker 4 (12:55):
Primarily, I'm curious because when I was reading or going
through your website and all of the content that is within.
Speaker 2 (13:02):
It, I did see that you refer to your services.
Speaker 4 (13:05):
As been design led, meaning I guess that you follow
a specific design methodology whenever you try to adapt that
low code solution to the business challenge.
Speaker 2 (13:18):
Right, I mean we Aspetrock.
Speaker 4 (13:20):
We also follow a few methodologies and we establish a
framework with a client where we try to, you know,
gather not only or set the expectations and gather insights
at the beginning, but we also exchange continuous feedback through
the development of the solution. Is this somehow related to
(13:40):
what you do when you're trying to implement that low
code solution based on what the business needs your client needs.
Speaker 3 (13:47):
Yeah? Absolutely yes. So what we mean by design letters
there is consumption at the root of everything that we do.
I mentioned that was the problem we set out to solve.
So consumption challenges solved by using story telling and information designed.
It's not important just to build a model which say,
can work on geospatial data and it runs AI algorithms,
(14:11):
but importantly you need to interpret it and for the
end user to recommended decision saying that this is what
you need to do and present it in an engaging
way using stories. So stories, when you say that is
a focal point, it cannot be an afterthought. You need
to have that as the front and center in terms
(14:32):
of where you start the discussion. So when we talk
to clients, we talk about what are the business priorities,
what decisions we need to enable and how we can
enable it for a particular user, and what are those
points that we need to how can we design that
So that becomes the starting point and the entire the
(14:53):
framework or the model we've built for delivery is inspired
based on that, and we have a a framework which
is inspired by Agile in terms of continuous cycles of development,
and we have over the last ten years we have
built our own methodology delivery methodology. Broadly, we call it
a radar framework, which is the overarching framework which covers
(15:19):
everything that we do at gram Now, one of the
key elements of that is, say, when it comes to implementation,
the design lead approach, storytelling lead approach. So there are
several toolkits to use under this broad methodology. That's how
we operated GRAMM.
Speaker 4 (15:35):
Very interesting, and I mean you've been in Grammar since
the ficking in because you're one of the founders, so
you most likely established the foundations of the initial delivery team.
You were mentioning that you use an Agile framework. But
since you adapt that low code solution to each one
(15:58):
of the cases that you find or that you encounter.
I have a few questions, do you I mean also
to compare your way of doing things against our way
of doing things and see what we can.
Speaker 2 (16:10):
Learn from each other.
Speaker 4 (16:11):
Do you have separate sprints for each client and account
because ourselves we do have a common sort of meeting
every week where we put together everything that is happening
across our portfolio. And then they have different squads for
each of the clients. Because we may puritude as fts,
(16:31):
some of our consultants data engineered data scientists that later
are part of the individual sprints that happen within the clients.
Speaker 2 (16:42):
What's your scenario like.
Speaker 3 (16:44):
As is inspired by a jet? We still have those
milestones which are closer to the traditional waterfall kind of
meteorology in terms of requirements design development. But in each
of these we have multiple multiple iterations that we run through.
For example, if when we're building this, there, when we're
building this. I talked about the initial design thinking workshop
(17:08):
where we try and sketch out what the solution should
look like or where should they even build, where should
they even apply data and analytics? So those are the
initial foundational discussions, and after that understand looking at the
data and understanding what is the right approach, how data
and analytics can be applied in this approach to solve
this business challenge. And then information design, how design can
(17:30):
come in, data visualization and machine learning algorithms can come in,
so all of those. If we talk about those initial milestones,
they we identify that and then they run iteratively. But
once we finalize that this is the approach and this
is the final deliverable likely to be, then that is
when we start multiple sprints in terms of running these
(17:53):
sprints to develop different versions of the application, so using
the lowcode platform, as there is a list of about
say sixty features to be implemented, so we have this continuously.
Every week, we have some features being developed. Initially it
could be a combination of some critical back end data
(18:14):
connection features, and then there would be a skeleton being
built a UI layer, and then there will be some
charts which will come in after that. So progressively you
see that being built and we have regular reviews of
this solution as it is being built with the client stakeholders.
So it's a combination of a child and some of
(18:34):
those standard milestones, just that we do it iteratively within
primarily when it comes to development and also in the
design phase there are iterations they run through.
Speaker 2 (18:44):
My question was only it related to how you deliver that.
Speaker 4 (18:49):
For a client, but also I mean you have different
projects that are being run in parallel and you have
a theme of how many data professional sort of close close?
Speaker 3 (19:02):
Two hundred and fifty is our team strengths?
Speaker 2 (19:04):
Wow? Congratulations?
Speaker 4 (19:06):
Well, first of all, congratulations, but if you have one
hundred something professionals there that are split across accounts, if
you say like that, do.
Speaker 2 (19:17):
They work not in insulation?
Speaker 4 (19:19):
But are they gathered by squads based on the challenge
the business challenges that they are facing and the solution
that they are building. And then do they have some
sort of unified or is there someone who has a
unified view. I guess there will be some similar role
to a CEO OH that has an overall view across
(19:42):
the whole portfolio projects.
Speaker 3 (19:45):
So, firstly, the unifying the process of the methodology is
there is a delivery framework which we follow religiously across
every single project and this is currently in the sixth version.
We launched the very first version. I think the second
year of operation. It's currently version six is going on.
So that is a mature framework depending on whether it
(20:09):
is a basic analytics like a diagnostic analytics project, or
a machine learning or a deep learning kind of a project.
Depending on the project, it is adaptable and we have
ways to execute the delivery framework, and we have account
leads people who oversee the relationship dot com and also
(20:33):
they have ownership for the solution the effectiveness of the solution.
So assuming there are five projects running for a client
and that person has that single point of view in
terms of how we are touching the different parts of
the business, how they all come in together, because often
there is there is synergy between these applications. So that's
how we have that ownership. And at the overall organization level,
(20:56):
when you talk about across all of these accounts, across
all of the client accounts we have been working with
about one hundred plus clients. Across all of these accounts,
we have a delivery head who brings in that the
complete organizational view to look at what is going well,
what is not going well, and how we need to
adapt our delivery framework to the new technologies which.
Speaker 2 (21:17):
Are coming in.
Speaker 3 (21:18):
So that's how we manage and keep it up to date.
Speaker 2 (21:22):
Okay, got it now? And based on your experience, I
mean now that you've hired.
Speaker 4 (21:27):
I don't know if you were involved throughout the hiring
process of all of those one hundred fifty something people,
but based on your experience, what makes great data scientists
a great data engineer? I know that being well at
being good at programming at Python or using I don't
know the tech stack of us here maybe useful. But
(21:49):
apart from that, what are the soft skills that you're
after whenever you are.
Speaker 2 (21:57):
Helping with the interview process.
Speaker 3 (21:59):
So there are five roles which we hire for and
have written about it in my articles. Broadly, it is
the data consultant role, functional data scientists, which is statistics,
machine learning, data engineer which is development operations DevOps and
programming role. And we have the design expertise information designers,
(22:21):
and we have project managers. So these are five key
roles we sta for and each of these, each of
this has a specific skill set and mixed coming into play.
I think based on the primary area I mentioned it
will be obvious. But to your question, what is that
most important skill we look for across all of these roles.
It is the application of technology to solve a business challenge.
(22:46):
So you might have someone who's great in deep learning,
but they are not able to understand the interpretation of
it to a business problem or how to apply and
solve solved for a specific scenario. Then that skill is
not put to the best choose. So we always look
for application skills, and that's a common challenge we've seen
in the industry. How you can bridge the technical skills
(23:08):
with what is needed on a project and can people
walk a few steps and apply their skill set. So
that's what differentiates from a great individual from someone who's
good enough or average in their skill set.
Speaker 4 (23:23):
I agree, But at the same time, what you're saying
is extremely hard and very unaffordable sometimes because you may
get someone coming out of university or has a few
years experience, maybe two or three, that is very good
at using technology and at understanding technology. But since hear
or see hasn't spent that much time facing business problems here,
(23:49):
see hasn't built that business document because it takes time,
doesn't it. So how do you get hold of this
talent before it gets too expensive.
Speaker 3 (24:01):
We have a mix of freshers, people from campus and
the latters. So when it comes to freshers, we invest
time in getting them on board and training them for
these skills so that we have good content we have
developed in the house, and we also put them through
very tailored shadowing projects and on the job training. So
(24:23):
we have figured out a way to use these people
and develop these applied skills for freshures. When it comes
to laterals, we're looking for that's something which we test
as part of the interview process to the best extension possible.
But that's something which we constantly if we see some shortcomings,
we again mentor and support people, but through a combination
(24:46):
of channels right when it comes to laterals, either through
assignments of practices or through scenario based questions. So that's
how we try and identify the fitment.
Speaker 2 (24:57):
I mean something that we do worse helps when we
try to do that. I mean, we we've.
Speaker 4 (25:03):
Built a close relationship here with the university and we
try to not much but link the academic world with
what's happening in business right because sometimes it's lacking behind.
I mean, the the programs around technology are not so
(25:26):
up to date, and we try to make sure that
you know everything that happens in university really allows the
students to hit the ground running whenever they join us.
So we created link between us and the university and
it really helps us to you know, build the specific
(25:47):
expertise in different technologies that that helps us.
Speaker 3 (25:52):
Yeah, that's a great point. In fact, I'm glad that
you're doing this. This is something that we have been
doing since that first year of inception as well. We
have partnered with several universities and universities across different disciplines,
whether it is the machine learning engineers, or data scientists
or even the design information designers. We have MOUs with
(26:14):
many universities and there are many where we have helped
them design the course, the curriculum and we run guest
lectures or in some cases we have run full like
full semester, handled the complete lecture. So we continue to
get involved. And this is across geographies in in India,
in the US as well, and we've seen that make
(26:35):
a big difference because often we go back and hire
from these campuses. We develop a relationship with these campuses
and hire students from there, so that also helps to
bridge some of these gaps even before a person comes
on board. That's a good point.
Speaker 4 (26:50):
Yeah, if I may, what's the split of stuff that
you may have based in the US or or even
around the world against India And where did the company begin?
Speaker 2 (27:02):
Did you start in India?
Speaker 3 (27:04):
Yeah, we started in India. For about five years we
were based exclusively in the Indian geography, and after that
we've branched out. Today we have presence in North America,
in the US, Canada, and Singapore, and so there are
a few other places where we're seeing greater interest for
our offerings. But if roughly, if we look at the split,
(27:27):
majority of the people, large majority of the people are
in India, I would say about roughly about ninety percent
of the teams in India and other global locations.
Speaker 4 (27:38):
And I guess that you open in your offices headquarters
for just locating people remotely in Canada, North America, and
all of that is because you find more business opportunities, right,
more commercial opportunities.
Speaker 2 (27:52):
There is a larger market to tackle.
Speaker 3 (27:55):
That's right, ya, closer to the client and yeah, h.
Speaker 2 (28:02):
Okay, I know.
Speaker 4 (28:06):
The next question that I'm going to ask would take
a couple of hours of debate, even though I'll try
to be very concise. If if if I were to,
I mean, as I said, we started better a couple
of years ago. Were started with a strong premise in mind,
(28:27):
which is, we want to make change and technology easier
for our clients by you know, using data science as
a tool that makes them operate better and more efficiently.
Because we felt that in the past, data science and
(28:47):
an AI was perceived as something very techy, very deep tech,
far from being business practical, and we wanted to change that.
So probably linking the world of the scientific technical experise
and what's providing ro O I to the business. And
I feel that that's something.
Speaker 2 (29:08):
That you've tried to do over the past ten years.
Speaker 4 (29:11):
So if you were to sure world learning with me
around your entrepreneurship.
Speaker 3 (29:19):
Past, which one would it be? I think that's a
good question. There's a lot which which can be added there.
But let me if you have to share just one point, right, Uh,
it's just uh, staying resilient and staying in the market.
So over the last ten years, we have seen a
lot of technology trends come and go, and the needs
(29:41):
of the clients also change. If you, if you stay
true to your mission and do not get swayed by
those short term either peaks or the trucks. For example,
in the early early years there were this there was
a strength around big data, and there was a lot
(30:04):
of investment going in that direction. And then suddenly there
was about five or five years back, AI enterprises started
investing a lot more. So many trends would come and go,
and you might see some short term business coming in
that direction. And when that trend it becomes a fat
and goes away, that business would no longer exist. So
(30:25):
there will be a lot of ups and downs. My
recommendation would be stay true to your mission and if
you have chosen a big enough problem, a beginnough vision,
that won't change despite whatever small the hypes come and go.
So stay true to that. And then if you're able
to stay in the market, because often in this space,
(30:50):
and that's the case with many of the businesses as well,
business to business enterprise selling, the cycle time is usually long,
and often you establish in relationship and eventually the project
will come in only six months later or a year later,
or in some cases it comes like three to four
years later. So we have seen cases where six years
(31:11):
after we made the first contact, a person reaches out
and says, hey, we were impressed by what you showed
six years back. Are you ready to do a project now?
So that staying in the market and staying proved to
your mission is what matters. That would be my recommendation,
Thank you very much.
Speaker 4 (31:27):
And based on that, I guess that something that you
said as a goal from the beginning is steering away
from only being perceived as a software service, plug and
play solution and being considered considered as a strategic partner
right in that data science sort of path at your
clients are brave or trying to get ready to take.
(31:51):
So how do you do this because it happens to
us right Sometimes it depends on the engagement and it
depends on the level of antuity of our clients. But
some of them they request a piece of development that
helps their data science.
Speaker 2 (32:08):
You need or it you need to move ahead when we.
Speaker 4 (32:12):
Really prefer, you know, as a service or professional services
company to build a long term relationships.
Speaker 2 (32:18):
So how hard was this for you?
Speaker 4 (32:21):
How did you ensure that you say you state in
that strategic path, the strategic positioning.
Speaker 3 (32:27):
That's a great question. Usually the projects might start transactionally
with with a small need or there is a burning
problem a client wants to solve there within the next quarter,
so that's how they reach out. And unless you are
a big established player in an industry, people may not
come to you for a long term relationship from the outside.
(32:47):
But what you'll have to use to your advantages when
these initial transactional or short term engagements come in. Take
it as a challenge to show enough value in that
so that you get into the radar of this client
and after that continuously keep up selling, move up and
talk see how you can MAC to the right stakeholders,
(33:10):
the right sponsors, and if you can demonstrate that you
can solve a bigger problem, show examples, show your past
case studies, and talk value proposition as opposed to solving
a transactional challenge. So if you're able to upsell and
demonstrate that big thinking and MAC to the senior stakeholders
and use that language, I think that steadily you will move up.
(33:30):
In some cases will be very rapid second third engage
you can get there. Others it might need some more time,
but you'll eventually get that that needs the focus that
you're very clear what problem you want to solve and
where you want to MAC. But you're patient enough to
work through the projects and work your way into that
high level mapping quick one.
Speaker 2 (33:51):
I know my answer, but I want to know yours.
Do you really date as high ends?
Speaker 4 (33:56):
Analytics are so powerful that it can transform many aspects
of our society. Society, I mean, I've seen that it can,
and I believe that it will continue to shape society
as a whole.
Speaker 2 (34:10):
I mean it's in everything.
Speaker 4 (34:11):
I think it's in autonomous driving, it's in our own devices,
it's in decision making. But to which extent do you
believe data science or advanced analytics If you want to
use the password AI because or it falls under the
umbrella somehow of data science you can too, to which
extent is going to change our world?
Speaker 3 (34:33):
I'm both excited and worried about the potential of data
analytics and broadly AI. It is very powerful. Having worked
in the space for over a decade, I see, I
see the potential, and at the same time, I'm worried
that it doesn't There aren't enough checks in the market,
and the incentives for taking it in the wrong direction
(34:56):
based on short term profits is too enticing for many
organizations and many individuals. So it is very powerful, but
it could post a big threat. So what I'm excited
about is there is a way to do this right
and this can be that invisible force which influences and
(35:19):
shapes everything that we do, whether it is our personal life,
how we lead our lives and how we make personal decisions,
or at an organizational level, how you run a company,
how you sell to clients, or even big ticket issues
like climate change and other environmental aspects. I think this
can play. This can play the role of a catalyst
(35:40):
and can be that invisible influencer to move things in
the right direction. But what we need to be clear
about is are we aligned on the right outcomes and
is it equitable fair? And we have some minimum regulatory
checks and self impost checks in place with that people
don't misuse it. So I think that I want to
(36:03):
see this evolved, and I want to see more checks
also come into place so that this can be sustainable
and it can really create a better future of everyone.
Speaker 2 (36:13):
Awesome.
Speaker 4 (36:13):
I'm glad to hear that you're taking those measurements sort
to say, I think they you In April published also
some guide guidelines, and I guess the US and different
parts of the world will continue to do so at
a government level. But I'm really glad that you bring
this up. Okay, yes, trying to wrap up. I think
(36:34):
it's been a very interesting chat with you. I really
enjoyed it also because I feel that I'm sure in
the same shues that you were in a few years ago. Again,
congratulations on what you've done with Graminar. I think building
a one hundred and fifty people business, it's something that
I do admire. Some last question, any recommendations on future guests.
(36:59):
I know that you that you've been part of dead
x talks, so probably you know a few interesting people
who should I v is picking with about the data
science and AI and the future of this practice.
Speaker 3 (37:11):
There are several interesting people, and that's one reason I'm
active on LinkedIn, because I see I bump into a
lot of great people, great content. I can definitely send
you a long list of people like I can shoot
that over to you or email. But there's a lot
of people doing good work. And I see also this
thought leadership thing taking taking shape where people are sharing
(37:35):
value and they are doing it selflessly. So there are
many interesting individuals to talk about. And in terms of
the earlier point you mentioned at Bedrock, I think I've
seen your work and you're moving in the right direction.
So very excited to see you grow. So good luck
in your journey and we're happy to have this conversation
(37:56):
with you.
Speaker 2 (37:57):
Awesome. Thank you so much. And last one read It
wasn't the last one. I need a book.
Speaker 4 (38:02):
I need something to read that you know inspires me.
Do you have any book in mind that you think
I should be reading in the next few weeks.
Speaker 3 (38:11):
I've been reading. The current book I'm reading is Competing
in the Age of Ai. Okay, professor his name is
Kareem Lacani. He's at Harvard Business School. That's an excellent
book which talks about building an organization in the current
age where you have algorithms pretty much taking over all
aspects of the business. So that's an interesting book. Gives
(38:32):
great examples of new age companies and how some traditional
and old companies are reinventing themselves. So that would be
my recommendation.
Speaker 2 (38:42):
Thank you very much. It's a great.
Speaker 4 (38:46):
I think it sounds really good. I need to get
that and I need to start looking at it. Okay,
thank you, Ganesh. I really enjoyed our call. I really
enjoyed recording this and hope the best for you and
for Grammar.
Speaker 3 (38:59):
Thank you, Thank you for Take It by Uri, play
like
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!