AI-Driven DevOps: Accelerating Innovation and Agility in Business Operations

CXOToday has engaged in an exclusive interview with Alok Uniyal, Vice President and Head of IT Process Consulting Practice, Infosys

1. How does leveraging AI in DevOps processes drive innovation and agility within businesses?

DevOps helps teams to build and deploy software solutions in an accelerated manner, using a highly automated tool chain, with built-in quality and security controls. AI helps in tapping into the power of the data that gets captured in the underlying tool chain and provides the following benefits:

• Provides insights to various stakeholders, across the value stream, based on real-time data analysis, which in turn drives faster decision-making and course correction, as needed, based on facts and figures.
• Predicts potential issues, which allows the team to take pre-emptive action, thereby driving greater agility.
• Amplifies the ability of the team to experiment and innovate based on data-driven insights and what-if analysis.
• Optimizes physical and/or cloud resource allocation/utilization and automates repetitive tasks – thereby driving greater cost efficiencies.

2. Can you elaborate on how organizations can implement DevOps across various technologies and contributes to accelerating an organization’s value stream?

In our experience, the best way to implement DevOps in a multi-technology environment is to set up DevOps as a common capability orchestrated by a centralized platform, which offers DevOps as a service to various teams across multiple technologies. Self-service reference pipelines (templates) for various technology stacks, completely baked with quality & security tools and gating, are created centrally, and made available on demand. Applications, across tech-stacks, can be onboarded onto the platform in waves. The Infosys DevSecOps Platform (IDP) is one such enterprise grade platform that allows enterprises to implement DevOps at scale.

In addition, it is critical to put-in place a measurement system that enables performance tracking, reporting and governance. DORA metrics is a good base to start with – Lead Time to Change, Deployment Frequency, MTTR and Change Failure Rate. Real-time visibility into the value stream based on these metrics goes a long way in driving acceleration and business outcomes.

Also, DevOps transformation is as much about people and mindset change as it is about process, tools & technology change. Given the pace at which the DevOps related tooling and practices are evolving, organizations need to invest in continuous learning of their talent as well as inculcate an agile mindset.

3. What challenges do organizations typically face when integrating AI with DevOps, and how does Infosys NextGen DevOps address these challenges?

Typical challenges are as follows:

• Security & compliance: Getting security clearance especially for open source LLMs, as there are concerns around privacy, security and intellectual property.
• People change-management: Even though AI Tools may be integrated, the cultural shift to embrace new technology is sometimes difficult. It requires enabling people as consumers of AI, articulating how it makes them more effective and productive.

Infosys NextGen DevOps leverages the AI-related assets, accelerators, and body of knowledge available as part of Infosys Topaz – the AI CoE, in addressing the above challenges. The Responsible AI framework, for example, guides teams in addressing the security / compliance related concerns.

4. Can you elaborate on how DevOps Transformation contributes to driving innovation and agility in businesses?

DevOps transformation contributes in the following ways:

• Validates business requirements quickly with faster and reliable deployment: New features are deployed to production at speed and at increased reliability with the automation of build, test, and deployment processes. The organization can iterate through their products faster and respond quickly to customer feedback.
• Amplifies productivity and quality: With increased automation using CI-CD, test and deployment automation, AIML assistance in development and testing and gated processes baked in the pipelines, the feature team can deliver high quality output and focus on innovation rather than repetitive tasks.
• Enhanced scalability and reliability: Usage of techniques like observability, chaos engineering, Infrastructure-as-Code helps in designing systems that are scalable and reliable. Such systems are able to cope up with the fluctuating business demand, experience little/no downtime, and provide a superior customer experience.
• Healthy work culture: DevOps processes ensure optimal flow of information among the team members and facilitate better communication and a transparent culture. With repetitive tasks being automated through the DevOps tool chain, teams feel motivated to work on higher-order items, innovation etc. A happier workforce is likely to be more productive.

5. How does personalization intersect with AI in enhancing customer interactions?

One of the top imperatives for business is to give tailored experience to the customers to boost engagement and loyalty. We are working on AI algorithms that provide real-time content delivery to customers based on preferences and behaviors. AI can also scour through large volume of customer data to provide various insights and tailor the interaction based on personal preferences and previous interaction history. There is also a degree of predictability with AI around personalized recommendations and content, increasing the likelihood of better customer engagement.