Robotic Process Automation (RPA) refers to the use of software bots for automation of business processes. RPA has found several use cases in HR processes and sub-processes which can help lean out operations in the form of lowered cost, significant efficiency gains, and higher quality output. In this report, we comprehensively analyse the HR process and identify the relevance and adoption of RPA across various sub-processes. It evaluates the adoption of RPA by specific use cases and assesses the limitations across each of them. The report also deep dives into the RPA talent requirement, skillset anatomy and emerging trends in RPA talent building.
The role of a Data Scientist is the most sought-after position in the world of tech. In today’s data-driven business environment, organizations are looking for data scientists who can turn volumes of data into actionable insights. The niche data scientist persona is often undefined and misunderstood which creates several problems throughout the hiring lifecycle. The average time taken to hire a data scientist in the US is roughly 48 days as opposed to the 37 days it takes to fill a software engineer’s position. Higher attrition rates (14% > 9%) and lower average tenures (1.5 years < 2.4 years) are also plaguing companies looking to hire and retain Data Scientists.
Most of these problems stem from the lack of clarity in how to identify the right Data Scientist and what companies really want out of them. Identifying and alluring the right person for the job is almost always more challenging than hiring itself because data science is a highly diverse and complex pool of skillset and talent.
While the Data Scientist usually gets assigned a multitude of names across various organizations, we’ve identified four broad types which are – Data Scientist, Applied Data Scientist – NLP, Applied Data Scientist – Vision, and Data Scientist Speech Recognition. The average Data Scientist must be aware of tools such as ML Frameworks (SciKit learn), DL Frameworks (Theano), and TensorFlow, Keras, Caffe. An Applied Data Scientist and Data Scientist typically possess skills like Gurobi, CPLEX, Symphony, HTK, TTS, and OpenFST. Here is a breakdown of behavioural skills, tool proficiencies and experience that each type of data scientist is expected to possess.
A Data Scientist is ideally proficient in programming languages like R, Python and OpenCL. Some of the other IT tools and platforms a Data Scientist must know are:
• Applied AI as a Service – Image Recognition AAS, NLP AAS, Vision AAS • Integrated AI Platforms – AZURE ML, AWS Deep Learning AMI, Google ML, DataBricks ML • AI Frameworks – SciKit Learn, TensorFlow, CAFFE, TORCH • Big Data Platforms – EMR, DataBricks, Spark • Infrastructure & Processing – GPUs, CPUs, AWS Lambda
Role of HR: Creating the Ideal JD
Creating an optimal JD attracts the best talent, and HR teams must cover all bases in order to get the right fit that goes beyond an individual’s skillsets. • Identify core skills required for the job – Statistics, ML, NLP or Computer Vision, specify the programming languages that are typically used, and define the understanding of frameworks based on the team’s tech stack. • List the responsibilities to be taken up by the employee across key business functions – Sales & Marketing, Finance, Operations or Customer Experience, and specify certain use cases the hire is expected to work on. Accentuating on key organizational values that a data scientist would prefer working in is another point to tick. • A culture that promotes learning and innovation, supports risk-taking behaviour, and a fast-paced environment will generally attract quality talent pool. Having a gender-inclusive environment, and using gender-neutral titles in the JD also helps. • Use descriptive titles like “engineer”, “project manager”, and “developer” as opposed to titles that suggest masculinity. Also, avoid using gender-charged words. For example, words such as “analyse” and “determine” are typically associated with male traits, while “collaborate” and “support” are considered more feminine.
At Draup, we take problem-solving to the next level by using cutting-edge technologies and having some of the brightest minds running highly complex algorithms on huge volumes of data. While we have the capabilities to solve most challenges, we also rely on open-sourcing which equips us with the right tools which allow us to find truly unique and innovate solutions to several problems.
The team at Draup has been trying to solve the problem of stateful deletion/updation at the ORM level for Django. Whilst solving the problem, we also wanted to build a generic solution for ORM so it could be used by Django-developers as well.
Stateful Deletion/Updation :
Stateful deletion is when an object cannot be directly deleted, without a prior prompt message (when deleting a single object entity, the prompt message gesticulates all the other dependent objects tagged along with the parent entity, before deleting it).
Stateful updation occurs while transferring object dependencies from one object entity to another object entity, through which duplicate entries from database-tables can be erased.
Draup Django :
Draup Django is primarily a Pip Package. It provides stateful deletion/updation and is, available here.
Installation :
pip install Draup-Django
Utilities :
Get Affected Objects :
Lists out all the dependent objects(Prompt-Messages).
getAffectedObjects takes input dict with id as key and reference of the model.
class parent(models.Model):
name = models.CharField(max_length=100)
class child(models.Model):
parent = models.ForeignKey(parent, on_delete=models.CASCADE)
name = models.CharField(max_length=100)
Parent Table :
╔═══════════╦════════════════════╗
║ Id ║ Name ║
╠═══════════╬════════════════════╣
║ 1 ║ Alice ║
║ 2 ║ Tom ║
╚═══════════╩════════════════════╝
Child Table :
╔═══════════╦════════════════════╦═══════════════╗
║ id ║ Name ║ parent_id ║
╠═══════════╬════════════════════╬═══════════════╣
║ 1 ║ Bob ║ 1 ║
║ 2 ║ Jack ║ 2 ║
╚═══════════╩════════════════════╩═══════════════╝
Code :
from draup_django import utility
m = models.parent
"""
Get Prompt Message
"""
error_list,prompt_message = utility.getAffectedObjects({‘id’:1},
m)
"""
Deletion with all dependent objects
"""
error_list = utility.deleteObject({‘id’:1,’force_delete’:True},
m)
"""
Transferring Object dependencies
"""
source = m.objects.filter(id=1).first()
destination = m.objects.filter(id=2).first()
error_list = utility.updateObjectDependencies(source,
destination)
Output : Prompt Message of getAffectedObjects function
[{'message': 'This object has been used in child 1 times.',
'model_name':'child','Parent models': '', 'count': 1}]
Note :
Every function returns the error_list, and hence only the operation is successful only if it is empty.
According to the above child/parent table, the updateObjectDependencies function will transfer the object dependency from id:1 to id:2, so the parent_id of Bob will become 2.
Updated Child Table :
╔═══════════╦════════════════════╦═══════════════╗
║ id ║ Name ║ parent_id ║
╠═══════════╬════════════════════╬═══════════════╣
║ 1 ║ Bob ║ 2 ║
║ 2 ║ Jack ║ 2 ║
╚═══════════╩════════════════════╩═══════════════╝
Limitation :
updateObjectDependencies will not work in case of a unique constraint, and one-to-one field in the model.
Feel free to share your thoughts and feedback to keep adding more such utilities. Reach out to us at info@draup.com.
About Draup :
Draup is an enterprise decision-making platform for global CXO leaders in sales and talent domains. Draup combines Artificial Intelligence with human curation to help organizations make data-driven strategic decisions. The platform is powered by machine-generated models, which are augmented by a team of analysts adding their learning-based insights to provide a 360-degree transactable view of their sales and talent ecosystem.
The highly specialized and often misunderstood nature of Analytics has organizations struggling to identify the right talent to create high-impact teams. The basic premise of creating an effective analytics team should revolve around recruiting individuals who are proficient in a specific area of analytics. You bring a range of uniquely skilled individuals together and allow the sum to become greater than its individual parts. This is where recruiters usually miss the bus as far as their new-age talent planning efforts are concerned.
Data scientist roles are quite fluid and have a way of fitting into every business function. Corporations are starting to make the shift from counting on business analysts and their intuition to data scientists and their cold hard data-driven insights. But this fluidity comes with a unique set of challenges along the way in terms of team structure and hierarchy. Finding the right balance or mix of data scientists and creating a formal reporting structure has been a problem plaguing startups and corporations alike.
Creating a data science team would require organizations asking themselves critical questions such as ‘how do I define their roles?’, ‘what is the formal team structure?’, ‘who do they report to?’ and ‘what business function do they live in?’. To help understand the convoluted world of AI and Analytics, Draup performed a study to understand the deeper characteristics of AI and Big Data roles and understand their impact on the global talent ecosystem.
The Data Scientist title has either expanded or has been used rather loosely and become all too vague. However, we look at some commonly accepted titles and what their roles encompass.
Data Scientists analyze raw data and turns them into actionable contextualized content.
Data Architect’s role is to capture, structure, organize, centralize, and contextualize data.
Data Engineer – These individuals lay the building blocks or infrastructure required to prepare and transform data in the organization.
Data Analysts – Their role includes gathering data, evaluating large datasets, data quality, develop analysis and reporting capabilities.
Draup’s Study on AI and Big Data
Draup conducted a comprehensive study to understand the deeper characteristics of AI and Big Data talent that can enable HR leaders in their workforce planning initiatives. We focused our analysis on top G500 R&D spenders, start-ups and service providers. We began with a corpus of around 28 million job descriptions which we narrowed down to around 1 million jobs that are most relevant as of 2018.
We identified 17 unique job roles within AI and Big Data/Analytics. This is made up of 7 primary Big Data roles, 3 primary AI roles, and 7 auxiliary roles. We’ve identified these roles across G500 corporations, service providers and start-ups to get an accurate mix of mature as well as new-age roles.
Talent Supply-Demand Analysis
The global supply of AI/Big Data talent is nowhere near meeting the prevailing demand and the imbalance is expected to go on for the next decade or so. To lay the groundwork for effective talent planning in new age skills, we performed a preliminary research which revealed that the global demand for AI/Big Data talent is close to 1.2 million jobs as of 2018. Within this, there is an unmet demand of around 500,000 and close to 300,000 of these jobs are open in the US alone.
Despite the high influx of talent in these emerging skills, there’s still a vast gap between the talent supply and prevailing demand. We’ve estimated close to 515,000 job openings in AI and Big Data/Analytics. US, China, UK, and India have the largest number of job openings in these skills with over 60% of these openings in the US alone. Globally, we’re set to see a huge boost in the number of job openings over the next few years. Job creation for AI and Big Data roles will reach close to 1 million by 2021 with an average CAGR of 23%.
Understanding the Data Scientist Role Progression
Organizations hire data scientists, data analysts and also data engineers to handle their complex data requirements. The role of the data scientist has evolved over the last decade and has funneled through from parallel roles such as statistician simply because of how these roles are intertwined. Draup analyzed the data scientist profile across the top 10 tech giants. We found that the business analyst and algorithms engineer roles are the key roles which have progressed into the modern-day data scientist role. Some other common progressions are illustrated below.
Concentration of Talent Across Verticals
From our preliminary analysis, we’ve identified the presence of installed talent across G500 corporations which are consolidated in tier-1 locations. They account for nearly 44% of the total employed Big Data and AI talent pool. Over 60% of the demand is distributed across enterprise software, consumer electronics and banking and financial services. Emerging global locations are also showing high talent scalability potential thanks to governments taking cognizance of AI and Big Data and increasing spending to strengthen infrastructure and incubating the ecosystem.
US, China, Israel, and India have a high concentration of AI and Big Data talent. The talent in China, Israel, and the US is predominantly employed by US-based tech giants while India has most of its AI/BD talent with service providers such as IBM, TCS, and Infosys. This is contrary to trends in Canada and the UK where there is a large presence of AI and Big Data talent at start-ups and niche sized mid-companies.
Global Hotspots and Emerging Hotbeds
There’s a high concentration of AI and Big Data/Analytics talent across hotbeds such as Boston, Seattle, Atlanta, Tel Aviv, Bangalore and Shanghai. The real talking point, however, should be the emergence of talent in Tier-2 locations. Several financial, socio-economic and bureaucratic factors have led to talent migration from Tier-1 to more conducive Tier-2 locations. Draup has identified over 130 emerging hotbeds for AI and Big Data skills globally. Over 20% of the total talent is currently located in Tier-2 locations and is only set to grow further. Our initial projections also indicate the presence of over 1mn machine learning developers across 37 countries by 2030.
University Talent Supply Assessment
Draup’s talent module analyzed universities in the US to identify the most sought-after institutions, and key courses in Artificial Intelligence, Machine Learning and Analytics. Stanford University has the maximum number of ‘Center of Excellence’ collaborations with tech companies. Carnegie Mellon University has emerged as a hotspot for AI and Big Data talent and has the greatest number of Machine Learning / Big Data courses and publications. CMU, UCLA, MIT and Stanford offer the most advanced courses such as Computational Biology, Genetic Algorithm and Cognitive Modelling Robotics. CMU has also collaborated with giants like Apple, Google, and Amazon for research in the field of AI, Robotics, and Deep Learning.
Geographic Analysis of AI/Big Data Talent
US – The US has the largest concentration of AI and Big Data/Analytics talent. Over 65% of the US talent is spread across Seattle and the Bay Area with Data Science being the most employed role across top firms. Nearly every engineering priority is focused on building cross-industry AI platforms.
·Seattle – Second largest talent hotspot in the US with majority of the talent in data management and data science roles employed across G500 corporations.
·Boston – High consolidation of AI and Big Data talent across startups in healthcare and retail industries majorly employed from Tier-2 Universities.
·Austin – High availability of niche talent employed across healthcare and retail.
China – Chinese tech giants Baidu, Tencent, and Alibaba have over 2,000 AI/BD engineers spread across tier-1 locations such as Shanghai, Beijing, and Shenzhen. There are 20+ universities that offer PhD programmes in AI/Big Data and over 30 universities offering undergrad courses in AI and Machine Learning. Overall, China has a very mature AI/ML talent ecosystem with Chinese technical universities producing ~80,000 fresh talent annually.
Israel – All eyes are definitely on Israel as they grow towards becoming a global AI and Big Data powerhouse. A big portion of the talent in Israel is spread across start-ups focused on industry specific applications such as cybersecurity and healthcare. Israel has over 400 startups in the space of cybersecurity which employs a big portion of the available talent.
“An idea that is not dangerous is unworthy of being called an idea at all.” — Oscar Wilde Draup hosted the 1st edition of its internal Hackathon challenge to give our teams a break from the daily grind (rut) and allow them to be at their creative best. It attracted widespread participation from our tech and non-tech ninjas. The idea was to allow teams comprising of coders, designers, business associates, psychologists and data scientists who were crazy enough to create something truly innovative. The hackathon was a 24-hour event and teams were expected to build their product within this window and present it to our in-house panel.
The Hackathon promoted the concept of idea conceptualization. The only real criteria were to have teams of 4 and to make sure the project was completed by 10 AM the next day. Teams were free to create any unique ensemble of talent that they felt added value to their project. In all, 13 different project ideas were conceptualized by 52 participants. We opensourced libraries and projects to help with Draup branding and to allow teams to create truly unique projects.
The D-Day
The ‘Hack-ateers’ came prepared armed with their gear, some even came prepared to pull in all-nighters. It was time to get down to business. With a strict 24-hour window, there was really no time to waste. 24-hours later, we were greeted by 13 different project ideas that could not be any more different from the other. Some wanted to automate processes, some teams focussed on improving efficiency, while others created something entirely new. Extra brownie points for teams with projects that were in line with our offerings on the Draup platform.
24-Hours Later
The madness came to an end 24-hours later only to be confronted by the difficult task of picking an outright winner. These creative powerhouses made sure that the panel members had a really hard time picking a winner with their unique projects and innovate pitches. Unfortunately, it’s in the nature of competition to have just one winner.
Some Notable Entries:
Invincible Hacker
The Invincible Hackers presented an impressively accurate resume filter feature. The job-description based resume filter would recommend the top-N resumes which are most relevant for a specific job description and allowed you to scout the best possible profiles for a given role. It completely automated the manual process of scouring through potential profiles.
Can’t be Draupped!
A Draup-powered mobile app right in the pockets of our customers had our panellists’ interests peaking. What’s more, the app was equipped with voice capabilities that allowed users to search for corporate executives from our Rolodex library for their sales enablement and hiring needs.
And the Winner is…
Fabricate ML on a Click
The winning entry focussed on making Machine Learning models accessible to everyone! The team at Fabricate ML on a Click, did just that. Regardless of your technical background, you can create a Machine Learning model with just the click of a button. The project allowed users to create these models with minimal effort and training in Machine Learning. If the idea wasn’t already impressive enough, single clicking or drag-and-drop interfaces were a big part of their training models.
Our Hackathon Statistics:
52 overall participants 13 teams 4 members per team-3 technology specialists, 1 business leader 24 hours to code 1 winner 0 losers ∞ possibilities
It was clear that the teams were driven more by their inner desire to create something truly remarkable than by the prizes. It is this very drive to build great things that binds us all together here at Draup. If you’re to looking to start or further your career in technology, have the right skill-sets and the ability to be self-driven, we have exciting opportunities for you. Head over to our Careers section for a list of available openings in front-end and back-end roles.
Do you find that you are more sensitive to a colour’s contrast than your friends are? Do you feel like you’re more aware of the tick tock or subtle changes in temperature? Don’t worry, there’s nothing wrong with you! Each one of us has a distinct way to relate to or understand the world around us.
VAK, which stands for Visual, Aural/Auditory and Kinesthetic, are different sensory perceptive modalities through which we understand our world.
For example, people who rely primarily on visual cues, tend to buy clothes depending on the colour combination or patterns, whereas people who emphasize on the texture and fitting would be relying on their Kinesthetic cues.
David A Kolb was one of the earliest learning theorists who found that individuals have a preferred learning style through their dominant sensory perception.
From a sales perspective, it’s highly recommended to keep your customers’ learning styles in mind to foster effective engagement and establish good business orientation.
Naturally, your next question is, “How will I know the customer’s learning preference?”. To answer this, it is important to look for certain cues in their demeanour and communication patterns. Some commonly observed patterns are listed below. Visual
Characteristic features of a person who predominantly uses the visual mode:
Fast talkers
Loud while in conversation
Use words and phrases such as ‘see’, ‘watch’, ‘point of view’, ‘focus’, etc., anything that draws attention to an image
Business orientation: They will be more receptive towards engaging when the interaction involves using visual aids such as charts, pictures diagrams, etc. They believe in first impressions and dress appropriately according to the situation. They might form an opinion on your product/idea based on the projections you present to them. In a nut shell, they will buy the concept if something visually appeals to them. Aural/Auditory
Distinctive features of an auditory person:
They are soft spoken and slow in their speech
They enjoy music, rhythmic beats
Use words such as, ‘listen’, ‘ring a bell’, ‘sounds good’, or ‘rhythm’, more often. They might frequently use fillers such as ‘ah’, ‘umm’, in their conversation
Easily distracted by noise
They are known to filter out outside noise by playing music to concentrate better.
Business orientation: They are good listeners and learn best when they listen to someone explain the subject to them.
They base their opinions more on what they hear than what they see, hence you can make good use of words to captivate them to your discussion and facilitate business. Kinesthetic
Striking features of a Kinesthetic individual:
‘Feel’, ‘Try’, ‘Experience’, are common words from a Kinesthetic individual
They are people who learn quickly by relying on the sense of touch
They like to experience and learn
They can be good at sports, performing arts or any task which involves physical movement
Business Orientation: They prefer trying out things before forming an opinion on them. They usually communicate at a slow pace. They respond well to demos or trial runs as this enables them to get a hands-on experience.
There are two other less-popular sensory perceptive modes:
Audio-digital, refers to those individuals who would like to “make sense” of a concept by reasoning out the rationale behind it.
Characteristic features:
They are very formal and conservative
They fear being illogical
Their voice is generally a monotone with few intonations
They appreciate concepts that have depth rather than relying excessively on the visual attributes
They need structure to understand and articulate things
They like to be looked at as dependable hence, they tend to verify and fact-check information
Reading/Writing - These are individuals who like to learn from their meeting sessions and written material. Some commonly observed aspects about them are that they enjoy reading and writing.
Business Orientation: To explain things or to convey a message to them, it’s best to have a written version of it for them to read.
It’s best to use all the above-mentioned strategies in appropriate context.
When you meet a set of people, it’s recommended to greet everyone with a handshake (of course to a limited audience size!) and provide a crisp presentation with pictures, data, handouts, and verbally explain the topic at a moderate pace.
Disclaimer - Do not generalize individuals by the presence of just one of the above traits. This will help maintain an unbiased approach.
The term Unicorns was coined by Venture Capitalist Aileen Lee in 2013
symbolizing the statistical rarity of such successful ventures. However, these
organisations have now become almost commonplace, reaching 220+ in number and
multiplying at incredible speed, a CAGR of 233% to be exact during the last 3
years.
The massive economic scale of the Unicorns club can be understood from their
combined net worth which has reached nearly $724 billion USD in 2017, just shy
of Netherland's gross domestic product (GDP). If Unicorns were an independent
nation, they would form the world's 18th largest economy!!
These organisations, being digitally native, structurally lean and
technologically agile have emerged across all industries designing, hacking,
and changing products, structures and business models of traditional
industries.
The top of the herd is dominated by some of the very big ones including the
Uber, Airbnb, Chinese Xiaomi, Didi and China Internet Holdings which are among
the fourteen decacorns, a term used for private venture-funded company’s worth
over USD $10 billion.
What are they doing with the money? Not a lot, it would seem.
Leveraging our DRAUP platform, we estimated that Unicorns spend on technology
reached a massive USD 32 bn, growing exponentially at a rate of 133% YoY during
the last 3 years.
Product development and Infrastructure Support are the major spend areas for
all Unicorns. Nearly 47 percent of the total technology spend is consolidated
on product development, being the key differentiator of superior product
experience and growth enabler for Unicorns. Technology spending on digital
areas such as Design, Data science and AI are amongst the core focus segments
for creating a personalised experience in consumer space and intelligent
products in enterprise space.
On the infrastructure-support front, the heavy technology spending is to ensure
robust scale that can support a large number of customer transactions, active
security requirements and analysing a large dataset of customers, products and
other stakeholders within and outside the organisation. However, being born digital,
nearly 95 percent of their infrastructure is based on cloud platforms which is
reliant on global external vendors. The spending on Cloud hosting, network
management and security & maintenance activities is a huge component,
comprising of nearly 53 percent of their total technology spend. AWS maintains
a significant market share lead, controlling nearly 65% total market among
Unicorns. Unicorns such as Snapchat has already announced a contract with AWS
for spending USD 1 bn on latter’s cloud services over next 5 years.
How do these hyper scaling organisations put up large technology capability in
a short while?
The answer lies within the unconventional route which has been adopted by
Unicorns compared to the traditional organisations.
During the growth stage Unicorns rely on infrastructure from 3rd party
platforms, open source OS, libraries and other platform providers.
Take the case of Airbnb which has built its backend infrastructure supported by
open source nginx platform and used AWS for hosting. It has also developed
capabilities through partnerships, acquisitions or acqui-hiring niche solution
providers such as Twilio for communication, LocalMind for navigation, Trooly
for identity verification etc.
As Unicorns scale, they bring critical capability inhouse in order to provide
the users with a seamless experience across the product stack. They also invest
in enforcing a comprehensive product control as well as an overarching
security, privacy and regulatory compliance.
Many Unicorns such as Uber has invested nearly half a billion on Mapping
project to build inhouse capabilities and plans to create its own detailed maps
for traffic patterns, pickup location etc. for its autonomous car project.
Cloudera is building necessary data centre infrastructure in-house and Airbnb
has developed a search and discovery algorithm to search 35k diverse property
listing on its website.
As they develop deep expertise in their technology stack, Unicorns sometime
opensource their capabilities. Airbnb has already taken a step ahead by open
sourcing its AirMapView API to co-innovate and experience data from real world
models. Its API enables interactive maps for devices with and without Google
Play Services to support multiple native map providers such as Google Maps V2.
Given their openness to work and readiness to scale their existing capacity in
emerging areas, Unicorns are good potential segment for Technology Service
Providers (TSPs) to collaborate.
So, how should the TSPs leverage this hyper evolving Unicorns ecosystem? Are
Unicorns a viable target segment for technology service and solution providers
to target?
We have already seen instances where TSPs have partnered with scaled Unicorns.
Outsourcing giants such as Wipro has partnered with Uber at its Bangalore based
map improvement group. Tata Technology Services has announced plans to work
with NextEV and Faraday Future to codevelop battery technology for Electric
Vehicles in Bay Area.
But what are the activities where Unicorns need help to develop capabilities?
Most of the Unicorns have small existing development teams focussed on core
product activities. These companies face challenges in scaling up their
existing workforce and infrastructure set-up to address potential exponential
growth. Non-core functions such as Dev-Ops, Customer Support and Q/A are the
areas where Unicorns lack capability and leverage third party solution
providers to help them scale.
However, TSPs need a proactive sales approach and thinking while planning
for collaboration with these hyper evolving companies.
Unicorns needs are dynamic and short lived. Secondly, it has been observed that
their decision-making is widely different from traditional organisations. The
Product ownership and responsibilities in many Unicorns is distributed across
new age stakeholders such as CMOs, CROs, CAOs, CInOs etc., unlike the
traditional organisations where decision making is confined to senior
stakeholders such as CIOs.
1. TSPs need to be proactive in selling and need to engage early with young
Unicorns to offer integration services for scaling latter’s core product
capabilities
2. This would require TSPs to develop capability in digital areas to explore
new age partnership in segments such as User Experience, API management and
integration services etc.
3. Decision makers in Unicorns have diverse roles across emerging markets,
products and business segments. TSPs need to establish deep connections with
new age stakeholders as compared to restricting engagement with only CIOs
4. Most of the Unicorns does not have global engineering presence. TSPs need to
engage through an on-shore delivery model to sell customized services by
understanding the technology stack and dependencies of the prospect. This would
require deep understanding of the pain points by working closely with the
in-house engineering teams. Many events such as large VC investment, M&A
and leadership change are some of the major triggers which could lead to a
potential outsourcing requirement for Unicorns. An event such as a senior
leadership hiring from a global outsourcing organisation could be a signal for
strengthening existing product capability through investment in newer
technologies or talent.
Draup’s Signal Processing tracks and interprets subtle and obvious changes in
organizations to recommend proactive actionable decisions for long term sales
strategy.
The traditional automotive industry has undergone a massive transformation with digital native technology providers across different industry segments playing a critical role. Smart Mobility, Autonomy, Connected Car and Electrification are the major technology-driven trends emerging across the industry.
The autonomous segment in the automotive industry is seeing a major upswing due to factors like reduction in cost of Transport as a Service (TaaS), lowered government regulations related to testing and operations of autonomous vehicles in the US and significant advancements in Machine Learning technology. The current Autonomous Vehicle ecosystem has been rapidly growing through a rich infrastructure of network, cloud & insurance providers enabling new age business models.
A DRAUP study estimates the Autonomous industry to be worth $87 Billion by 2020 and enable potential savings of $2.2 Trillion in the areas of fuel efficiency, cost of life and productivity gains.
Digital Transformation of the Automotive Industry
“With increasing focus on digital engineering within Automotive, OEMs are prioritizing initiatives around connected and autonomous vehicles”
Around USD 8 billion is the Technology spending by the top 25 players in Autonomous vehicle segment. In-house R&D is focused on developing core software capabilities, leveraging deep learning for computing, vehicle control and vision-based perception.
The automotive industry, in general, and the autonomous vehicle segment in particular, is moving toward a new customer value proposition. Today, digital platforms are at the core of how customers figure out how to get from point A to point B. This makes it imperative for automakers to digitally transform their internal operations, service models as well as external partnerships.
While the automotive industry has always used information technology to achieve efficiency and scale, it is ripe for a digital disruption since the New age customer now uses a smartphone, is social media savvy and environment-conscious, and has become much more demanding in terms of speed and convenience.
This has led to the entry of AI / IoT native platform providers like Google and AImotive with strong AI capability that leverages deep learning algorithms required to make advanced driving systems safe and predictable. OEMs now have strategic focus on developing critical safety and driving systems in-house. OEMs such as Daimler, BMW and Ford are establishing partnerships with technology providers to collaboratively develop software capability for vision and perception systems.
Semiconductor giants such as Intel and Nvidia have developed specialised Autonomous Vehicle SoCs for processing and computing large amount of vehicle datasets using Machine Learning algorithms. While Traditional suppliers such as Bosch and TomTom have enabled advanced vehicle navigation and monitoring through specialised telematics equipment, new age suppliers have built capability into Advanced vehicle control using deep learning, sensor systems and connectivity services.
Well-funded start ups like Nauto, Argo AI and Drive.ai are the top players investing in full stack-Autonomous Vehicle solutions.
Changing Industry Structure
“Digital firms have disrupted the automotive sector across the value chain and as a consequence R&D focus has shifted to software”
In the traditional industry structure, Tier 1s and Tier 2s worked together to provide the Full-Stack of automotive solutions to the OEMs. This is now changing, with Tier 2s disrupting the traditional supplier relationship model to position themselves as a direct Full-Stack supplier of AV solutions.
For instance, Semiconductor giant Intel was traditionally a Tier 2 supplier who created ML/Deep Learning based microprocessors and sensor chips, and depended on Tier 1s like Continental and Bosch for sensors like LiDAR and RADAR, and 3rd party integration. In the new age industry structure, Intel directly offers full stack solutions related to autonomous vehicles – cameras, in-car networking, sensor-chips, roadway mapping, cloud software, machine learning and data management, to OEMs.
Intel accelerated its AV push through acquisition of Mobileye, bringing later’s core AV capabilities inhouse. Intel has now positioned itself as a full stack system integrator across vision systems, sensor technology and computing platform, and has also partnered with companies like Harman for its connectivity platform and Velodyne for its LiDAR technology.
The industry is also seeing more partnerships such as the one between Intel, Ericsson and GE to launch an open industry platform – 5G Innovators Initiative.
Challenges and Opportunities for Traditional Automakers
The automotive business is undergoing a transformation, leaving turmoil and uncertainty in its wake. Traditional automakers are dogged with concerns such as declining car ownership, flat revenues, new competition from non-automakers like Google and Uber, a new ecosystem of OEMs, suppliers and dealers, and changing government regulations. Auto sales are seeing a slowdown in all major markets including Europe, US and Japan.
This presents both challenges and opportunities for traditional automakers. In addition to investing more on profitable segments, culling declining segments and expanding into emerging geographies, it has become essential for OEMs to invest in “riskier” new age technologies such as autonomous and connected vehicles, ride sharing, electric vehicles etc.
DRAUP estimates the Engineering Services market in the Autonomous & Connected Car segment alone to be $215 Mn – $225 Mn.
A lot of automakers like GM, Toyota and Daimler are investing in ride-hailing companies, which act as a great platform to test their autonomous prototypes on.
Ford is an example of an automaker who was struggling with declining stock prices, flat revenues and decreasing profits. Ford was seeing low sales of their “Car” product line and declining growth rate even in their top geography, Northern America, due to a change in the customer mindset. This led to an R&D spend of $7.3 Billion in 2016, an aggressive increase of 9% over the previous year, on high growth areas such as Autonomous, Connected and Electric vehicles. Ford realigned its vision and strategy in 2016 to evolve from being an automotive company, to being an automotive and mobility company. While Ford is spending $4.5 billion to expand its EV portfolio by 2020, its investment of over $1Bn in AI, LiDAR, 3D mapping technologies and talent hiring from Silicon Valley, puts it ahead of the others in the Autonomy Landscape.
“Automotive firms will continue to gear up investments around autonomous capabilities as it becomes more mainstream “
Analyst firm IHS Markit predicts that autonomous cars will reach global sales of 600,000 by 2025 and 21 million by 2035. It is becoming apparent that those auto companies who focus on R&D spending and patents in non-traditional Business Units, realign their engagement strategy with the ecosystem, connect with new age digital suppliers and invest in engineering talent in new age technologies, will be ones who will stay competitive in the years to come.
“An idea that is not dangerous is unworthy of being called an idea at all.” — Oscar Wilde Draup hosted the 1st edition of its internal Ha...
The highly specialized and often misunderstood nature of Analytics has organizations struggling to identify the right talent to create hig...