Data

Embracing AI in Agile teams

We connected with QA’s subject matter expert Michael Easson, to capture his top insights into the impact of the AI revolution on the world of Agile.

AI is transforming team dynamics and productivity. It isn’t just an extra tool; it’s changing the way we work, prompting us to rethink our practices and accountabilities.

1. Rethinking team structures

Agile development has thrived on small, cross-functional teams delivering high-quality products. Now, AI can act as a highly knowledgeable team member, enhancing a team’s expertise and efficiency.

Imagine a team member who excels in every programming language and framework. This is how AI significantly boosts capabilities. According to Michael, “a developer skilled in prompting AI can achieve results in hours that previously took days.”

So, while cross-functional teams remain essential, the need for overlapping human skills is diminishing. A team of one or two, augmented by AI, can access an array of knowledge and skills. For instance, GitHub Copilot, an AI-powered code completion tool, can help write code more efficiently.

2. Smaller teams, better collaboration

AI can handle many routine tasks and provide vast knowledge, enabling smaller teams to achieve more. An ideal team size with AI partners is three to four members. This setup balances the benefits of small, agile teams with the enhanced capabilities of AI.

Spotify uses AI to analyse project data and predict risks, allowing teams to address issues proactively. Smaller teams with AI support can focus more on strategic, creative, and problem-solving activities, fostering innovation and rapid iteration. This reduces dependencies and streamlines workflows, accelerating delivery times.

However, more teams mean an increased need for cross-team coordination. Effective project management and inter-team communication tools are essential for alignment. Michael’s prediction is that “instead of reducing the workforce, smart companies will AI-empower everyone, transforming larger teams into multiple smaller, more efficient units.”

3. Transforming change review boards

Consider an AI agent designed to replace the Change Review Board (CRB). An AI-empowered team might interact with this AI like this:

  • Request review of a feature update for compliance and potential impacts.
  • Receive AI feedback on compliance with coding standards and regulatory requirements, and recommendations for additional testing based on pasts issues caused by similar updates.
  • Implement further tests and request a simulation of the effects of this on the current system.
  • Receive AI estimation of impacts, such as increased load time, and recommended solution, for example optimising database queries related to the new feature.

This speeds up the review process and improves accuracy, freeing the team to innovate. Google’s AI-driven code review tool, Code Review Bot, automates parts of the code review process, significantly reducing the time developers spend on reviews.

4. Evolving role of software engineers

AI’s ability to generate code is changing the role of software engineers. They must now oversee architectural decisions, create precise prompts, review AI-generated code, and ensure quality. This allows for focus on strategic decision-making over manual coding, transforming engineers into mini-product owners.

Change emphasises the need for reskilling. Engineers should focus on learning AI system management, ensuring ethical standards, and tackling complex problems that AI can’t solve. Human creativity and strategic thinking can then complement AI’s strengths.

Companies like IBM have already launched extensive reskilling programmes to train their workforce on AI and related technologies, in preparation.

5. Improved cycle time

With AI speeding up development, traditional Scrum sprints might shift towards shorter, dynamic cycles. Instead of bi-weekly cycles, we could see daily teams' syncs, goals, and review progress.

Organisations first need AI enablement at all levels. This means:

  • Data integration: Seamless data flow enabling AI to make real-time decisions.
  • AI tools and training: Equipping teams to use AI effectively.
  • Continuous feedback loops: Implementing systems for AI to continuously learn from data and user interactions.
  • Flexible infrastructure: Adopting cloud solutions and scalable infrastructure to support AI operations.
  • Collaboration platforms: To facilitate communication among smaller, agile teams.

Accenture has adopted AI tools across its Agile teams to shorten development cycles and improve productivity. AI has allowed them to move towards continuous delivery.

6. ‘Floating’ specialists

Human specialists remain essential but in a new capacity. AI can handle routine tasks, while complex or advanced issues still require human expertise. These specialists might not belong to a single team but float between teams as needed.

A prime example is the evolving responsibilities of scrum masters and agile coaches. Instead of primarily teaching and mentoring, they will pivot toward coaching and facilitating.

AI-empowered teams can independently access vast amounts of information, reducing the need for instruction. Coaches and scrum masters will guide teams in effectively using AI tools, to foster creativity, maintain performance and offer strategic insights.

7. Reskilling over downsizing

Another pointer from Michael: Don’t downsize, invest in reskilling.

According to McKinsey, 82% of executives believe retraining and reskilling are at least half the solution to skills gaps. BCG highlights that the average half-life of skills is now less than five years, making continuous learning and reskilling critical.

Reskilling can help retain valuable employees and enhance productivity. IBM’s reskilling programmes have shown that employees who train in AI and related fields are better prepared to handle new technologies and contribute more effectively.

With the disruption of rapid advancements in AI and automation, reskilling ensures that employees remain relevant and capable of working alongside new technologies, to maximise productivity and innovation.

Moving Forward with AI

As we embrace AI, it’s essential to reassess practices. Michael advises to “reflect on current processes, identify necessary changes, and adapt to this new reality. Embracing AI is not just about technology but about transforming how we work to stay ahead in an evolving landscape.”

To be ready for transformation, focus on reskilling rather than downsizing. Continuous learning and adaptation will help maintain productivity, foster innovation, and retain talent.

“It’s time to drive forward with a commitment to growth and development,” Michael reminds us, “leveraging the power of AI to build a stronger, more agile future.”

Don’t know where to start?

Check out ICAgile Foundations in AI, to learn how AI can support your skills and creativity so employees can deliver work more efficiently.

  • Understand the evolving state of AI
  • Develop prompting techniques
  • Discover the strength of AI aligned to your business strategies.