Topics In Demand
Notification
New

No notification found.

Building Impactful Solutions With AI Developer Services
Building Impactful Solutions With AI Developer Services

317

0

 

       1.  Building Speed – The need for scalability and automation in AI

Tech teams need to build prototypes rapidly, push applications into production environments and speed up deployment. The question of success hinges on the aspects of access and scalability. This is true for all technologies, including Artificial Intelligence. On the one-hand, wider adoption of the cloud accelerated by the Pandemic, and even otherwise, provides the much-needed access and scalability, and on the other side, ready-to-use tools to automate the time-consuming and repetitive aspects of Machine Learning model development is helping reduce the time-to-market. The emergence of Low-code-No-code eases the adoption by teams that have limited Data Science and/or programming skills.

The widespread availability of ready-to-use tools to build and deploy ML models is expected to drive the adoption of Artificial Intelligence, powered by the Cloud. NASSCOM’s recent report ‘AI-as-a-service – Democratizing AI For Scale’ proposes, among other factors, that Cloud AI Developer Services is a key pillar within the scope of ‘as-a-service’ model for AI/ML which is witnessing traction globally. Going forward, this will open newer opportunity areas for companies trying start their AI journeys as well as for service providers with wider capabilities and global alliances.

Gartner defines Cloud AI Developer Services as:

Cloud-hosted or containerized services/models that allow development teams and business users to leverage artificial intelligence (AI) models via APIs, software development kits (SDKs), or applications without requiring deep data science expertise.”

The Cloud AI Developer Services market is fairly concentrated at the top. The four Big Tech companies, of which 3 are also Hyperscaler cloud infrastructure providers (Amazon, Microsoft, and Google apart from IBM), constitute the bulk of the market. They have well-defined, wide, and deep portfolios of platform and infrastructure (also known as Cloud Infrastructure and Platform Services – CIPS), and a mature global ecosystem of alliances, not to mention the huge data repositories they own.

       2.  The current scope of Cloud AI Developer Services: (non-exhaustive)

Sl. No.

 

 

Functional Areas

What is included?

Market-leading services

1.

 

 

Language

Text-to-speech, Speech-to-text, Sentiment Analysis, NLP/NLU/NLG, Text Analytics

Azure Cognitive Services, Azure Machine Learning, Microsoft Conversational AI, Google Document AI

2.

 

 

Vision

OCR, Video Analytics, Image Recognition, Vision Intelligence

Visual Inspection, IBM Maximo Visual Inspection, IBM Watson Knowledge Studio, IBM Watson Discovery, Microsoft Cognitive Services, Azure Machine Learning designer

3.

 

 

AutoML (Automated Machine Learning)

Automation tools to help in data preparation, feature engineering, model creation, deployment, model monitoring and model management

Amazon SageMaker Autopilot (SageMaker Studio, Data Wrangler, Clarify etc.), Google TensorFlow, Google Kubeflow, Google Cloud AutoML

 

 

 

 

 

 

 

 

 

 

Gartner’s 2021 Magic Quadrant Report for Cloud AI Developer Services, puts AWS, GCP, IBM and Microsoft as global leaders in this space. These 3 and IBM have multiple offerings for each of the above categories – Language, Vision and AutoML. 

 

Gartner Magic Quadrant for AI Developer Services, 2021

     The Rise of AutoML (Automated Machine Learning)

A large chunk of these services falls within the category of AutoML or Automated Machine Learning. The other categories of AI Developer Services fall within the realm of Language and Vision. These two categories will be taken up in subsequent articles.

“Automated Machine Learning provides methods and processes to make Machine Learning available to non-Machine Learning experts, to improve efficiency of Machine Learning and to accelerate research.”

www.automl.org

    1. Off-the-shelf AutoML tools

There are many off-the-shelf AutoML tools available on the market that can facilitate large-scale adoption of Artificial Intelligence across business functions, spanning industries. Some of the popular tools are Auto-PyTorch, TransmogrifAI and auto-sklearn.  Machine Learning-as-a-Service (MLaaS) is another closely related aspect that makes adoption of Machine Learning easier by enabling users to jumpstart their journeys without having to make large-scale investments. MLaaS, also known as Machine Learning Cloud services, leverages the computing power of the Cloud and is one of the newer consumption models.

    1. AI Professional Services

Gartner predicts that by 2024, more than 50% of all organizations would have leveraged AI service providers for AI consulting, implementation/managed services. Gartner thus defines the market for external AI service providers as a subset of the broader market for data and analytics (D&A) service providers, encompassing consulting, implementation and managed services for AI techniques, which may also include AI governance, security, audit and monitoring.

Data and AI Key to Achieving a US$5 trillion Economy

      Evolution of AI-as-a-Service in India

The AIaaS market in India broadly follows the same evolution pattern that is visible globally. The market is fluid and is still developing with a multitude of players and participants constantly evaluating their offerings on the market and bringing up new features, products and/redefining/positioning them. Within the scope of the current document, the market for AI Developer Services in India is mature. This segment is led by global Cloud Infrastructure/Service providers – AWS, Azure, GCP and IBM, as has already been highlighted above. 

NASSCOM’s recently launched report AI-as-a-service – Democratizing AI for Scale postulates that the three top reasons driving deployment of AI at-scale in India are:

  1. Lack of in-house capabilities to launch AI
  2. Flexibility offered (in AIaaS) in terms of cost and resource allocation
  3. Availability of off-the-shelf tools which can be accessed via APIs and SDKs.

Building expertise in the areas of data preparation, creating, deploying, monitoring and managing models, and buying compute capabilities can be expensive, long-drawn or fraught with dangers of failure. AIaaS enables businesses to leverage the benefits of Artificial Intelligence using cloud-hosted or containerized models via APIs, SDKs.

The India model has certain advantages which are primed at democratizing the adoption of AI. One of them is the highly mature GCC (Global Capability Centers) ecosystem in India. India has more than 1,430 GCCs which are rapidly transitioning into Global Centers of Excellence (CoEs) for their global parent organizations providing them with a host of AI services – MLOps, Model Governance Frameworks, Data Services (Data-as-a-service, labelling, annotation), ML Model Management. In the forthcoming articles, we shall try to delve deeper into the developing India story for AI and will look at how Data and AI hold the promise to deliver a US$1trillion data economy by 2025.


Follow me on LinkedIn https://www.linkedin.com/in/myurlformyprofilebandevghosh/


That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.


images
Bandev Ghosh
Senior Manager

© Copyright nasscom. All Rights Reserved.