Paving the paths to AI engineering and ModelOps

Posted on Posted in IBMBigData

In the face of tremendous disruptions and shifts during 2020, IBM Institute of Business Value (IBV) uncovered that organizations have continued to accelerate their digital transformation. According to Gartner, “Despite the global impact of COVID-19, 47% of AI investments were unchanged since the start of the pandemic, and 30% of organizations actually planned to increase AI investments.”[1] Indeed, McKinsey found that respondents attributed as much as 20 percent or more of their organizations’ earnings before interest and taxes (EBIT) to AI. Further, in a recent survey on 3,000 CEOs, IBV found three natural segments: 48% of CEOs are customer-experience focused, seeking real-time feedback and engagement fueled by technology such as AI; 30% of CEOs prioritize innovation for products and services; and 20% of CEOs focus on operations with an emphasis on ecosystems.

As these AI adoption patterns vary by industry, company goals or operating conditions, you need an AI strategy that can scale across any patterns, including customers, products and operations. How can you exploit your technology investments more fully and demonstrate higher returns on AI investments? What’s worth looking into is AI engineering, one of the Gartner Top Strategic Technology Trends for 2021.[2]

What is AI engineering and why should you care?

AI engineering is a discipline of driving value through the application of AI, focusing on developing tools, systems, and processes. AI engineering provides a framework to design AI systems that demonstrate results in the real world. A robust AI engineering strategy will facilitate the performance, scalability, interpretability and reliability of AI models while delivering the full value of AI investments.

AI engineering cuts across the core functions of DevOps, ModelOps and DataOps to feed AI models into cloud-native apps for continuous integration and continuous delivery (CICD) of fresh, relevant data. AI engineering is ideally practiced by integrating with the existing DevOps frameworks rather than introducing a new set of isolated projects. The good news is, because they are already a key contributor to DevOps, developers are becoming a major force in AI. They are acquiring data science skills and actively engaging in AI development. Let’s look at the following use cases to learn how AI drives value for business when deployed at scale.

Raise the bar for customer experience with AI predictions

Excelling in customer experience is not just about addressing complaints or recommending what to buy next. “Experience bar-raisers” are applying design thinking,  a human-centered design practice, to anticipate scenarios and what customers need, mine feedback for deeper engagement, and improve business readiness through the experience of employees and partners. Companies can better serve customers when they use AI to further accelerate these meaningful interactions—anticipating responses more accurately and automating mundane operational tasks. They can develop AI using a unified approach for automating AI life cycles through building, running and managing models. Learn more about how scaling AI at Lufthansa has helped avoid delays, better predict boarding times and avoid long queues at check-in counters.

Accelerate the innovation pipeline for your products

If you have a focused strategy on product-driven growth, you can give heightened attention to innovation pipelines. Through massive structural changes and behavioral shifts, industry leaders have been steadfast in turning new technologies into offerings. A prime route to lasting success is combining a data-centric approach with human-centric design and product and service excellence. A unified approach, bringing cross-functional teams together and building an automated end-to-end pipeline with extensive data ecosystems, fuels better predictions and optimizations for developing more innovative products. Learn more about how AI models were used to enhance offerings by reading “Wunderman Thompson: AI reimagined at scale.” 

Increase operational scale with ecosystems

An operations-focused leader prioritizes AI-driven automation, extended reach with partner ecosystems, and proactive measures to address security, compliance and privacy risks. When predictive insights and optimized logic are integrated into the application workflow and available for people to act on, you can get faster, higher returns on AI and app investments. Further, given the drastic change in global business and human behaviors, leaders with an operational emphasis are retuning or revamping AI models to be sufficiently performant for deployment with apps through DevOps. Through continuous model monitoring, you have an AI and operational system that can harness time-sensitive insights to drive efficiency and save costs in a risk-intelligent way. When you have an end-to-end AI development visibility, you can mitigate bias, drift and risk associated with operationalizing models. Read AI in urgent times at Highmark Health to learn more.

Get started

With IBM Cloud Pak® for Data, you can put AI engineering to work and address a variety of use cases with your own strategic focus – be it for customer experience, product innovation, or operational agility. Bringing data and AI capabilities together on an intuitive, integrated platform, you can build a ModelOps practice that can synchronize with DevOps and is underpinned by DataOps. You can make technology operations more agile and impactful while reducing the costs and risks of bringing AI and DevOps operations together. With its extensive ecosystem, IBM Cloud Pak for Data helps you extend your open source and third-party investments and tap into IBM’s latest AI innovations.  As part of building your AI engineering your practice, you can retire technical debt and modernize your information architecture to be future-ready.

Next steps

Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.

[1] Smarter with Gartner, 2 Megatrends Dominate the Gartner Hype Cycle for Artificial Intelligence, 2020, Laurence Goasduff, September 2020.

[2] Smarter with Gartner, Gartner Top Strategic Technology Trends for 2021, Kasey Panetta, October 2021.

The post Paving the paths to AI engineering and ModelOps appeared first on Journey to AI Blog.