Top data science and AI trends to watch out for in 2021 | AIM & AnalytixLabs

Top Data Science & AI
Trends To Watch Out For In 2021
This report by
highlights the top
trends that will
define the industry
in 2021
December 2020

Disclaimer: This document is the result of research carried out by AIMResearch. Permission may be
required from AIMResearch for reproduction of the information in this report. Reasonable efforts have
been made to source and present data that is believed to be reliable, but makes no representations or
warranty, express or implied, as to their accuracy or completeness or correctness. All rights reserved with
the aforementioned parties.
© 2020 Analytics India Magazine Pvt Ltd and/or its affiliates. All rights reserved. Images or text from this
publication may not be reproduced or distributed in any form without prior written permission from
Analytics India Magazine. The information contained in this publication has been obtained from sources
believed to be reliable. Analytics India Magazine disclaims all warranties as to the accuracy, completeness
or adequacy of such information and shall have no liability for errors, omissions or inadequacies in such
information.
This publication consists of the opinions of Analytics India Magazine and should not be construed as
statements of fact. The opinions expressed herein are subject to change without notice.

Introduction 05
Preface 04CONTENTS
EmergingTech Will Fill Essential Gaps Across
Non-Traditional Domains 15
NLP Models Will Be BiggerThan Ever And Power
Human-Machine Conversations 13
Automation And Intelligent Machines Will Drive Critical
Roles 11
Cloud Will Back Analytics Deployments 08
AI In Cybersecurity Will Address Major Security
Loopholes 18
Businesses Will See Real-Time Actionable Data
Consumption 20
Customers Will BeThe Priority 22
Federated Learning & Counterfactuals Will BeThe Game
Changer 24
AIOps, MLOps & ITOps Will Streamline Processes 26
Policy & Regulatory Interventions Will Define
Production Of ML Models 28

Top Data Science & AI Trends To Watch Out For In 2021
04
PREFACE AI is a powerful catalyst, and having multiple vaccines for the
novel coronavirus in less than a year is the most significant testimony of this in
present times. In the 200-year-old history of immunization, never before such a
feat has been achieved before this.
Amid the unprecedented challenges 2020 posed, Data Science enabled tech has
touched every sphere of our lives like never before. This pandemic has put the
adoption of digitization on steroids, which, in turn, is going to feed AI to grow
more pragmatic. The year 2021 is bound to carry forward this momentum,
tapping many unexplored opportunities and make AI even more pervasive in our
lives.
In this well-researched report, esteemed industry leaders have shared their take
on key trends that will shape the AI & Data Science industry’s future. I believe
this to be very informative and useful for Data Science enthusiasts and aspirants.
Sumeet Bansal
CEO & Co-Founder, AnalytixLabs

By Analytics India Magazine & AnalytixLabs
05
The year 2020 was full of unexpected challenges. Having said that,
it also served as a unique opportunity to leverage technology on multiple fronts.
From adopting it in various industries such as retail, eCommerce and others,
to adopting it to ensure the safety of employees in work from home scenarios,
and improving consumer experiences, the industry went through various dig-
ital touchpoints. Adoption of data, analytics, AI, cybersecurity and other new
technologies saw an exponential growth to bring about changes to fit into the
changing business scenario.
Looking at the previous year, 2021 looks like an opportunity for tech trends to
grow to newer arenas. Intelligent machines, hybrid cloud, increased adoption
of NLP, and overall an increased focus on data science and AI is going to be the
highlights in the coming year. Some of the other trends that may see a rise in the
coming year are pragmatic AI, containerisation of analytics and AI, algorithmic
differentiation, augmented data management, differential privacy, quantum ana-
lytics, among others. Considering these trends, it can be said that data is increas-
ingly becoming a critical part of organisations after the pandemic.
The annual data science and AI trends report by Analytics India Magazine aims
to highlight the top trends that will define the industry each year. This report,
which has been developed in association with AnalytixLabs, covers the trends
that will shape the year 2021. AnalytixLabs is a leading Applied AI & Data
Science training institute in India. These trends are a culmination of popular
industry opinions.
INTRODUCTION

Trend 1
Cloud Will Back
Analytics Deployments
Trend 2
Automation And Intelligent
Machines Will Drive Critical Roles
Trend 3
NLP Models Will Be
BiggerThan Ever And Power
Human-Machine Conversations
Trend 4
EmergingTech Will Fill Essential Gaps
Across Non-Traditional Domains
Trend 5
AI In Cybersecurity Will Address
Major Security Loopholes
Top
Data
Science
& AI
Trends
2021

Top
Data
Science
& AI
Trends
2021
Trend 6
Businesses Will See Real-Time Actionable
Data Consumption
Trend 7
Customers Will BeThe Priority
Trend 8
Federated Learning &
Counterfactuals Will BeThe
Game Changer
Trend 9
AIOps, MLOps & ITOps Will
Streamline Processes
Trend 10
Policy & Regulatory Interventions Will
Define Production Of ML Models

08
1. Cloud Will Back
Analytics Deployments
Cloud tech and edge computing will be central to all CIOs’
strategy as they gear up to manage the business team’s
demand for increased agility and near real-time response.
Cloud AI engineers who can deploy AI models at scale on the
cloud and integrate them with IT systems will see significant
demand.
Srikanth Velamakanni, Co-Founder, Group Chief Executive & Vice-Chairman,
Fractal Analytics
‘‘
To enable a robust remote workforce, organisations will
require the agility to adopt new technology. Building new IT
infrastructure on top of legacy systems within data centres is
no longer the solution. In this era, clouds will lead the way.
As such, it is paramount that security remains at the core of
this new technology adoption amidst a new wave of remote
workers.
Rohit Sawhney, Systems Engineering Manager, Juniper Networks India
‘‘
In the coming days, analytics will move to the cloud. While
non-critical and departmental analytics have moved to the
cloud, current investments will surely accelerate making cloud
as the mainstay. This will be facilitated by on-demand
compute, server-less analytics, elastic separation of compute.
Muraleedhar Ramapai, Executive Director, Maveric Systems
‘‘

09
By 2022, cloud services will be essential for 90% of data and
analytics innovation. Data and analytics leaders need to pri-
oritise workloads that can exploit cloud capabilities and focus
on cost optimisation and other benefits such as change and
innovation acceleration when moving to the cloud.
Nishant Rathi, COE and Founder of NeoSOFT Technologies
‘‘
The use of cloud technology for analytics rather than trans-
actional purposes only will gain currency. The year 2021 will
witness the reconfiguring of the cloud by scaling up memory
and improving the network bandwidth to facilitate real-time
analytics. The deployment of cloud-based AI is expected to
increase 5x between 2019 and 2023.
Rohit Manglik, Founder & CEO, EduGorilla
‘‘
The need for an agile cloud ecosystem has become evident
with the pandemic. A hybrid cloud enables companies to
achieve superior adaptability during unprecedented events. It
creates a confluence of private and public clouds via pro-
prietary software to facilitate seamless interactions between
different services. This year, we observed multiple large public
cloud providers developing tools to drive seamless connectiv-
ity between on-premises servers and the cloud infrastructure.
The reason why a hybrid cloud is more powerful is that it
gives businesses enhanced control over their data while also
making their operations more flexible.
Suhale Kapoor, EVP and Co-Founder, Absolutdata
‘‘
Continuing the momentum from 2020, we will witness an
increased momentum towards the adoption of hybrid cloud
services. While most organisations have adopted the new nor-
mal of working remotely, there will be an uptake in the adop-
tion of virtual office solutions and virtual cloud recordings for
conferences to enhance real-time engagement. With the kind
of data deluge that enterprises have witnessed in 2020 and
will continue to do so in 2021, adoption of a hybrid cloud
model for smooth orchestration of data management policy
from the edge to the cloud will be one of the main agenda of
decision-makers while going into 2021.
Ranga Jagannath, Director, Agora
‘‘

10
The technology of hybrid cloud data warehouses should see
the light of the day in 2021. Organisations storing their data
into multiple warehouses by different cloud providers based
on the specific use cases, with the ability to cross-query, join
and analyse the data in a single, optimised interface resulting
in cost-and-time-efficient analysis, will become popular.
Ankur Sharma, Head of Analytics, Instamojo
‘‘
Cloud adoption has increased multifold. The Indian
traditional sectors like healthcare, manufacturing, education,
etc., are also evaluating different technologies like AI, ML, for
achieving operational efficiency. As consumption increases,
production increases, lowering cloud subscription cost i.e.
economies of scale. This would benefit start-ups/ industries to
expand their IT sprawl more efficiently.
Sachin Waingankar, Head of Cloud, Web Werks Data Centers
‘‘

11
2. Automation And Intelligent Machines
Will Drive Critical Roles
Marion Roussel, Head - Decision Sciences at Capital Float
‘‘
Biswajit Biswas - Chief Data Scientist, Tata Elxsi
‘‘
Kaustubh Chandra- Director – Marketing & Digital Sales Group, NTT Ltd. in India
‘‘
Technologies such as Robotic Process Automation (RPA) will
help automate low-value efforts, including data collection and
processing, providing a quicker way of using data to focus on
high-value activities such as data analysis and modelling. Data
points such as GST, POS terminals may be sourced to provide
insight and opportunities to fintech organisations and help in
analysing risk, fraud detection, thus creating tailored made
offers.
Business Process Automation has got a tremendous impetus
and top-most priority in any CXO’s list, given the unique
circumstances we went through in 2020. Intelligent automa-
tion is about augmenting this with intelligent decision
making and filling the crucial gaps for some hard to replace
HIL (Human in the loop) based systems. Intelligent
automation adoption will see cross-connecting advancements
of different technologies like virtual conversation agents, NLP
based entity engines, computer vision with edge AI, and large
scale bot orchestrations.
Our research shows that 70.5% organisations cite improved
CX as the top factor driving their digital transformation. We
foresee greater adoption of chatbots and AI-driven natural
language processing bots which will increasingly undertake
businesses’ early-stage interactions with customers. Auto-
mation will also play a critical role in employee experience
initiatives. Here, we can expect to see advances in and adop-
tion of robotic process automation, machine learning and AI.

12
Mike Tria, Head of Platform, Atlassian
‘‘
Digital products are exploding all around us. Yet many of
them are quite static, non-personalised and non-intelligent. AI
and analytics-enabled intelligence that is embedded in these
products will need to become aforethought by including them
in the product architecture and MVP1 planning and not rele-
gated to subsequent release as an afterthought.
Nitin Sareen, Consumer Analytics Head, Wells Fargo
‘‘
Smarter AI will be the trend. I think that humans, at their
basic cell, are lazy. We want to make our lives simpler and
faster. And AI is going to be the major catalyst for that. We
are already using AI to do so many things at the tip of our
fingers and we will be doing so much more. AI will automate
the process of screening candidates, interacting with employ-
ees on the company portal and guiding them to FAQ answers,
streamline data, and present only useful information to the
users, and so much more.
Ankit Kashyap, Co-Founder, Vasitum
‘‘
Data maintenance and governance are key pieces of a compa-
ny’s data architecture. These are automation tools that help
data scientists speed up the process of untangling from data
issues. Recently, there has been a rise in machine learning
and artificial intelligence that has allowed for the creation of
new data science tools that automate a variety of repetitive
tasks. Machine Learning automation courses such as PyTorch
(+542%), TensorFlow (+458%) and OpenCV (+300%) have
seen an enormous spike.
Irwin Anand, MD, Udemy India
‘‘
Automation will be inevitable in 2021. As more engineering
organisations migrate their infrastructure to the cloud and
adopt a microservices architecture, the next challenge they’ll
face is figuring out how to manage and keep pace with an
unwieldy, distributed system. Whereas automation has been
seen as an emerging tool in the past, it will become essential
for survival in 2021.

13
3. NLP Models Will Be Bigger Than Ever
And Power Human-Machine
Conversations
Prasad Rajappan, Founder & CEO, ZingHR
‘‘
Rajesh Murthy, Founder Architect & Vice President (Engineering), Intellicus Technologies
‘‘
Vijay Ramaswamy, Partner and Co-Founder, Analytic Edge
‘‘
The year 2021 will be known for artificial intelligence, Nat-
ural Language Processing and predictive analytics as promi-
nent technology trends. NLP can be implemented in various
verticals from retail, HR tech, to healthcare to BFSI, and much
more. It can also improve the customer and employee experi-
ences, reduce the redundant activities for the people and help
to focus on value-added activities.
The rapid adoption of digital voice assistants like Alexa, Siri
has led to the need to access all information by voice com-
mands. NLP-powered BI is going to be a key 2021 trend
where a user could interact with its business data on a mes-
senger app, get predictive insights by simple voice commands
and thus effectively share it with the relevant stakeholders.
AI and ML tools and capabilities will increasingly be lever-
aged in various stages of the analytics process – from quality
control of data to identify incorrect data and catch outliers
during the data capture process. It will help build more robust
marketing analytics models faster, automate reporting and
insight generation using Natural Language Processing (NLP)
and Natural Language Generation (NLG), among others. All
of this will eventually translate into quicker turnaround times,
lower costs and sharper insights for clients.
4
8

14
Dinesh Sharma, Co-Founder & CTO, AskSid
‘‘
Cognitive bots will become the ‘de facto’ virtual assistant
of future knowledge workers. What makes this technology
powerful is that users can ask questions in natural language
and expect a meaningful response from the AI engine. AI algo-
rithms in 2021 will increasingly become popular in answering
standard FAQs as well as help lawyers, doctors, loan officers
with all kinds of answers.
Narendran Thillaisthanam, VP - Emerging Technologies, Vuram
‘‘
Multimodal integration of visual and language processing
can solve a number of complex problems in multiple fields.
Capabilities like video KYC, fraud & forgery detection and
Document AI will be strengthened to their optimum by Deep
Vision & Language Integration.
Virendra Kumar Pal Chief Innovation & Analytics Officer, SMEcorner
‘‘
With bigger than ever state-of-the-art language models cre-
ating a huge buzz with their releases in 2020, NLP has now
gained more attention. Many more retail brands shall now
tend to use AI-based market and consumer insights in the
company’s decision making and marketing strategies. They
will adopt Conversational AI systems to generate consumer
provided first-touch data to know what their consumers need
and more importantly ‘why’!

15
4. Emerging Tech Will Fill Essential Gaps
Across Non-Traditional Domains
Dr Sheela Siddappa, Chief Advisor - Data Science, Bosch Engineering & Business Solutions
‘‘
Saurabh Awasthi, Data Scientist, Mondelez International
‘‘Data science is no more a support function but a requirement
across functions and domains. Data scientists have always
been in high demand, but Data science as a skill set is slowly
becoming a preferred requirement for business leaders. In or-
der to bring the transformational change and make data-driv-
en decision making a habit, CEOs are looking for business
leaders with data science skill sets.
So far, AI solutions were able to address problems within
limited scope/ scenarios /fluctuations in demand. The year
2020 has paved a path to develop AI solutions that will be
flexible yet robust to account for unforeseen circumstances/
demand and supply. Thus, the year 2021 will look for holistic
solutions to incorporate any odds that may occur. The biggest
challenge is to develop solutions with very limited or no data.
This would in-turn call in for Innovative Data Scientists who
are great at the fundamentals and can give a probabilistic
view to the decision, as the scenarios or cases considered will
be high. Solving such large-scale problems would take much
longer time, expecting advancement in infrastructure and Big
Data Architects to design more effectively.

16
Sreekanth Menon, VP - Data Science, Genpact
‘‘
With technology taking over the work-ecosystem in 2020,
there has been a steep demand for a future-forward subject
like Data Science within the country and beyond. The said
vertical has been evolving more than ever and has been rated
as one of the most exciting career paths of all time. Such
evolving circumstances have led to a wide range of new-age
job roles across the board.
Arjun Mohan, CEO - India, upGrad
‘‘
Rakesh Jayaprakash, Product Manager, ManageEngine
‘‘
Analytics will no longer be restricted to traditional functions,
i.e., banking, retail, transport, but become integral to everyday
operations. Data literacy, i.e. understanding and quantifying
performance using data will be indispensable to all roles/jobs.
Data-oriented training and upskilling will become crucial, and
leading indicators will replace lagging indicators to facilitate
informed proactive decision-making.
Atishay Jand, Delivery Manager, TheMathCompany
‘‘
To support demand shifts from POCs to scalable models, the
AI/ML community will dedicate efforts on full technology
stack enabled ML pipelines for continuous monitoring and
re-training of AI/ML models in production environments.
Emerging roles such as ML Engineers, Citizen Data Scien-
tists and Decision Scientists will be critical to achieving this
endeavour.
Now that organisations realise the effectiveness of analytics
platforms in measuring productivity, its usage is expected to
penetrate not just to IT but other departments that have
traditionally not relied on analytics applications. For
example, data from HR applications combined with employee
performance metrics, can be used to measure the influence of
employee welfare initiatives on employee productivity. This
focus on analytics brings an exciting dynamic to the way
organisations work and also introduces a set of challenges in
data governance and maintenance.

17
Shantanu Bhattacharya, Chief Data Scientist, Locus
‘‘
Dileep Mangsuli, Executive Director, Siemens Healthineers
‘‘
Arpita Sur, Assistant Vice President at Ugam, a Merkle Company
‘‘
Location intelligence will get a lot more prominence.
COVID-19 has fast-forwarded the digital penetration, and
now customisation of products to specific geographies is going
to be much more efficient based on data.
There will be special requirements for AI solution providers in
healthcare. All procedures used to maintain health, diagnose,
and treat diseases place the highest demands on trust,
integrity, reported benefits, and associated costs. These de-
mands will also apply to AI in terms of capabilities, quality
assurance methods, clinical validation, requirements for the
data used, the avoidance of bias, and clear measurement for
performance assessment.
Prashant Rao, Senior Manager, Mathworks
‘‘Creating successful AI-based systems require model lifecycle
management, which includes training, deploying, monitor-
ing and updating the model for the system. To achieve this,
in 2021, engineers will augment their workflows to include
development best practices and production pipelines with
IT. These will support the AI-enabled systems in real-world
environments.
The pandemic has pushed everyone to embrace remote com-
munication. This trend has been witnessed in healthcare too,
which has seen a rise in Telemedicine. The rise is expected to
continue in 2021. AI can help in the area of telemedicine in
various ways such as virtual assistants providing reminders
for medication, answering health queries, routine checks,
remote patient monitoring and management, and more.
Kunal Aman, Head Marketing- SAS India
‘‘If 2020 was the year of COVID, 2021 will be the year of the
vaccine. Analytics will not only play a critical role in clinical
trials and speeding the vaccine development process but will
also be vital for planning rollout and tracking distribution,
side effects and overall efficacy.

18
5. AI In Cybersecurity Will Address
Major Security Loopholes
Laurence Pitt, Global Security Strategy Director, Juniper Networks
‘‘
Geo-political and economic realignment of the global stage
could result in more intense cyber warfare. While cyber-
attacks would become more severe and frequent, resulting in
massive data thefts and service outages; cybersecurity would
need to play a fast-paced catch-up game. AI and ML would
play a critical role to detect such attacks at early stages and
prevent them from spreading virally. Fire Sale (of Die Hard)
scale of attacks can soon be a reality.
Gaurav Kumar, Founder and Chief Technology Officer (CTO), Valyu.AI
‘‘
For several years now, cybersecurity has been one area where
investment and budget growth are constant. The security team
has positioned successfully with insights and future trends,
and the business sees strength in security as both a regulatory
need and a competitive advantage. However, in 2020 we saw
a change: investment had to be brought forward to support
remote working, and a rapid move into cloud-based software
services, all driven by the pandemic. Now, 2021 may see
reduced spending on security, and an increased need to
demonstrate fast value from previous security investments.

19
Stephen McNulty, President, Asia Pacific at Micro Focus
‘‘
Jyoti Prakash, Regional Sales Director, India & Saarc Countries, Splunk
‘‘
In the face of security risks from an increasingly remote
workforce, organisations will increase investment into access
security, analytics, and automation to protect sensitive
information. Failing to cover end devices with rigorous
security policies has proven to be costly, and many
organisations have paid the price for that this year. As the
attack surface continues to expand in 2021, we can expect
more organisations to keep a tighter rein on intra- and in-
ter-organisational data flow, with defence measures
encompassing context-based access controls, geofencing of
employee remote work location, and encryption. Security
analytics and automation will become mainstream to help
organisations detect anomalies in user behaviour and deploy
quick remediation to block malicious activities.
2020 was a year of unprecedented challenge for IT security
teams. Technology is changing quickly, and an organisation’s
attack surface is constantly morphing. They have tools for
detecting anomalies — and detect so many that they’re
overwhelmed. Pandemic workforce disruption will drive a
greater focus on endpoint security and the zero trust model.
Faster-moving digital transformation will include more
artificial intelligence in the SOC and much more.

20
6. Businesses Will See Real-Time
Actionable Data Consumption
Kunal Kislay, Co-Founder and CEO, Integration Wizards Solutions
‘‘
In spite of recent advancements, much of the data gathered by
businesses remains under-analysed, unstructured and unman-
ageable in volume. Actionable data will be the name of the
game for companies that want to gain crucial competitive
advantages, and to obtain actionable data, Data Scientists and
Analysts will be in high demand.
Nikhil Barshikar, Founder, Imarticus Learning
‘‘
Reports are delivered real-time. The mobile app, tablet, web
or through PA system announcements, data consumption is
multi-channel and personalised. The output of algorithms is
no longer fed into age-old dashboards to generate reports. It is
increasingly delivered through a channel that makes the data
actionable. The growth of NLP, AR and VR have transformed
data consumption.

21
Sameer Dixit, General Manager, Data, Analytics, AI and ML at Persistent Systems
‘‘
Nidhi Pratapneni, Senior Vice President - Product, Analytics & Modelling; Marketing;
Public Affairs, Wells Fargo
‘‘
As we all get overwhelmed with reports, it is very hard to
figure out how to deal with these reports. Significant
bandwidth has to be spent in gathering insights from these
reports and datasets. We will see an increasing number of
solutions powered by ML, which will be able to ingest these
reports/datasets and autonomously generate insights in
natural language from them. These insights could then be
plugged into a low aperture device like Alexa to be read out
to the users.
Mundane as it may sound, the discipline of data operations
and all that it entails will be key to the efficient organization
along with use of data. We’ll see an agile approach enabling
a technological and cultural change that drives enhanced
collaboration and automation. A successful build-out of data
structures, with enhanced layers of protection, will become
the mainstay of a successful analytics organization enabling
reduced time to insights and decisioning.

22
7. Customers Will Be
The Priority
Deepak Pargaonkar, VP - Solution Engineering, Salesforce India
‘‘
The flexibility of choice provided to the end-consumer is
going to emerge as an essential feature. Every conversation,
every interaction point could be seen to carry options.
Emerging out of this theme would be the choice of response
channel, transaction channel, product forms, sales force
flexibility. Once a consumer makes a choice, effective delivery
of that chosen option will be the game-changer.
Moumita Sarker, Vice President – Client Delivery, Cartesian Consulting
‘‘
In 2020, bots are widely being deployed in customer service
apps, but will soon spread to sales management. As we
approach 2021, customers will increasingly demand
companies to deliver seamlessly. Despite the pandemic,
customers still want to receive the same, in-store, white-glove
experience as their online experiences. Companies will use AI
to deliver those experiences customers want. We expect to see
more virtual try-on, smarter delivery and pickup options and
interactive virtual event marketing.

23
Manish Jha, CIO, Addverb Technologies
‘‘
Zabi Ulla, Sr. Director - Products & Solution, Course5i
‘‘
We predict most of the new business systems will be using
continuous intelligence in coming years. Industries could use
continuous intelligence to monitor and optimise scheduling
decisions and also provide more effective customer support.
E-commerce companies are increasingly integrating AI in
various applications to gain a competitive advantage in the
market. The adoption of AI-powered tools helps them to anal-
yse the catalogue in real-time to serve customers with similar
and relevant products. This improves both sales and customer
satisfaction. E-commerce companies are also. Therefore, the
increasing application areas of AI in e-commerce is expected
to boost the growth of the market during the forecast period.

24
8. Federated Learning & Counterfactuals
Will Be The Game Changer
Radhakrishna Venketeshwaran, Head of SDC & Vice President - Product & Engineering,
Blackhawk Network India
‘‘
Satheesh Kumar V, Director – Technology, Synechron
‘‘
As the competition gets tighter and fiercer, product companies
have to be extremely innovative in winning the trust of the
customer and digital experience is the key to that. One way
for data science to pave the way for this aspect will be tapping
the Federated Learning Model and algorithms to capture Mo-
bile behaviour of customers. This will enable the creation of
personalised recommendations by deploying learning models
on the phone rather than on the cloud. It will ensure right
product sensitivity and relevance at the right time, resulting in
a win-win situation for both business and customer.
Federated Learning is an improved model in machine learn-
ing, enabling data scientists to perform ML on a decentralised
data and then aggregate (federate) the learning in a central
point for further learning and application. This trend will
catch a rage in the FinTech sphere as it would allow banks to
develop tailor-made solutions to a certain set of customers, by
practising ML on their premises and then aggregate the learn-
ing to develop plausible solutions. About the fashion in which
data is protected in FL, and when it’s coupled with differential
privacy controls, it will only gain popularity in multiple folds
in years to come, and not just in 2021.

25
Ramprakash Ramamoorthy, Product Manager, Zoho & ManageEngine Labs
‘‘As it currently stands, many AI models’ decisions are solely
made based on correlation, much like the decision-making of
a three-year-old child. However, these models are progressing
to include causal techniques and counterfactuals. As an ex-
ample, many banks use AI-models to make lending decisions.
One bank’s AI model may reject a loan application due to a
low credit score. However, a better model would account for
causal techniques and also enable counterfactual inference
over the loan applicant’s data points. In essence, the model
will show what other variables the applicant can include to
receive the loan. Additionally, causation and counterfactuals
help to prevent preconceived biases from arising from the
datasets. The incorporation of counterfactuals will become
increasingly prevalent in algorithmic decision-making.

26
9. AIOps, MLOps & ITOps
Will Streamline Processes
Dr Jai Ganesh, Senior Vice President & Head, Mphasis NEXT Labs
‘‘
Pankaj Kitchlu, Systems Engineering Director (India-SAARC), Juniper Networks
‘‘Next year, AIOps will go from theory to practice for many
organisations. With the increase of remote workers and the
home becoming the new micro branch, AI will become table
stakes for delivering a great client to cloud user experience
from client to cloud while controlling IT support costs for
remote employees. IT teams will need to embrace AIOps to
scale and automate their operations. AIOps cloud SaaS will
turn the customer support paradigm upside down. Instead of
users submitting tickets to IT, AI will proactively identify users
with connectivity or experience issues and will either resolve
(the self-driving network) or will open a ticket with suggested
remediation actions for IT.
MLOps (Machine Learning + IT Operations) focuses on
automating and productising machine learning algorithms
through enhanced collaboration and communication between
data scientists and information technology professionals.
MLOps empowers data scientists and app developers to help
bring ML models to production faster and at scale. The
objective is to accelerate the life cycle of machine learning
development, deployment and productionising of ML algo-
rithms. It will enable organisations to improve the quality &
reliability of the machine learning solutions in production and
helps automate, scale, and monitor them.

27
V Satish Kumar, CEO, EverestIMS Technologies
‘‘In the next few years, industry estimates indicate that over
55% of data will be created outside data centres or cloud
environments. This will be used in tandem with new applica-
tions at the edge to enhance operational efficiencies providing
impetus to enterprises. AIOps will be the key driver of this
forward momentum. It is the application of artificial intel-
ligence to IT operations which helps accelerate data-driven
decision making. The AIOPS approach will become crucial for
managing and monitoring IT environments that are complex,
hybrid and span distributed regions. It also helps in avoiding
Imminent Service Disruptions while facilitating Root Cause
Discovery.
Arpita Sur, Assistant Vice President at Ugam, a Merkle Company
‘‘AIOps helps automate IT operations by combining big data
and machine learning. As businesses rethink the new normal,
we’ll witness the rise of AIOps, especially for the prevention
of cyber-crime and automatic resolution of IT problems. In
2021, this will be ever more critical as a majority of the work-
force continues to work from home using multiple, and often
unsecured devices.

28
10. Policy & Regulatory Interventions
Will Define Production Of ML
Models
Sri Viswanath, CTO, Atlassian
‘‘
Mathangi Sri, Head of Data - GoFood, GoJek
‘‘
In the next five years, increased data and privacy regulation
will have a big impact on the way we design AI/ML models.
As a result, investment in data management is going to be
critical in determining the success of AI systems. Companies
that have better data management frameworks, platforms and
systems will win in building effective AI tools.
We have directly moved from an era of human-led data
decisions to machine-led data decisions. With the increasing
focus on privacy and data protection from regulatory bodies,
there is a need to have machine led AI to be guided by human
governance. We would be in the coming years going to see
the increase of more governance in the AI solutions. The next
phase would be machine-led and human-guided AI solutions.
Algorithms would no doubt be able to build patterns, but it
will also get governed by human empathy. Possibly we could
call this “Sensible AI.”

29
Pranav Bhaskar Tiwari, Programme Manager, The Dialogue
‘‘
Anand Sri Ganesh, Chief Digital Officer, BRIDGEi2i Analytics Solutions
‘‘
The pandemic has had an unprecedented impact on
adoption automation and industry 4.0 technologies with
sectors like HealthTech and EdTech being revolutionised like
never before. AI proctored exams are a commonality today.
The Government has already sought comments on the Indian
AI stack. While the Personal Data Protection Bill, 2019 is still
pending before the Joint Parliamentary Committee, policy and
regulatory interventions are expected to minimise the friction
as AI interacts with the fundamental right to data and
informational privacy along with that of free speech and
non-discrimination. This will be complemented with more
start-ups providing innovative solutions to secure user rights
in the digital space.
As the dependence on AI systems increases, the blurring of
boundaries between data governance and data privacy will
create ethical dilemmas. Responsible AI will be a reigning
theme for years to come, as enterprises and regulatory services
look for the elusive balance in managing inherent biases that
creep into algorithms and impute human behaviour.

About Analytics India Magazine
AIMResearch provides rigorous, objective research and advisory to organisations that plan to achieve
higher levels of success with their analytics implementations. Our single, overriding goal is to equip
our clients with the insights, advice and tools they need to create a well-oiled, data driven enterprise.
Visit us at www.analyticsindiamag.com
Or contact us at info@analyticsindiamag.com

About AnalytixLabs
AnalytixLabs pioneers in analytics training since 2011 and as one of the first analytics training
institutes, it is widely acclaimed and known for high quality training by industry experts themselves.
After establishing ourselves as the top analytics training institute in Delhi NCR, they slowly and
steadily progressed to earn the same reputation pan India based on their stellar record and student
satisfaction. Their students are placed in leading companies across industries like Accenture, American
Express, AbsolutData, Axtria, Bank of America and McKinsey. They are focused at helping their clients
develop skills in basic and advanced analytics to enable them to emerge as “Industry Ready”
professionals and enhance their career opportunities. It was co-founded by Sumeet Bansal, Ankita
Gupta and Chandra Mouli.
Visit us at www.analytixlabs.co.in
Or contact us at info@analytixlabs.co.in

www.analyticsindiamag.com www.analytixlabs.co.in
Analytics India Magazine
#280, 2nd
Floor, 5th Main,
15 A Cross, Sector 6 , HSR layout,
Bengaluru, Karnataka 560102
Get in touch with us at
info@analyticsindiamag.com
AnalytixLabs
Bengaluru | Gurgaon | Noida
Get in touch with us at
www.analytixlabs.co.in
info@analytixlabs.com

Top data science and AI trends to watch out for in 2021 | AIM & AnalytixLabs

More Related Content

What's hot (20)

Similar to Top data science and AI trends to watch out for in 2021 | AIM & AnalytixLabs (20)

More from Srishti Deoras (8)

Recently uploaded (20)

Top data science and AI trends to watch out for in 2021 | AIM & AnalytixLabs