Frank Bignone, SVP and Head of Corporate Strategy & Growth at FPT Software – Interview Series

Frank Bignone, Senior Vice President and Head of Corporate Strategy & Growth at FPT Software, brings more than two decades of international leadership experience spanning digital transformation, artificial intelligence, aerospace, geospatial technology, and enterprise growth. Throughout his career, he has held senior leadership roles across Asia, the Middle East, and Europe, including serving as Chief Digital Transformation Officer for FPT Software’s operations in Japan and the Middle East, and previously leading digital transformation initiatives across Asia Pacific and China for Airbus. His background combines deep expertise in AI-driven business transformation, strategic partnerships, market expansion, and large-scale technology adoption. Having built ecosystems across aviation, software, and enterprise technology sectors, Bignone has consistently focused on helping organizations leverage emerging technologies to drive operational efficiency, innovation, and sustainable growth. Today, he leads FPT Software’s corporate strategy and growth initiatives, guiding the company’s global expansion and next-generation service development.

FPT Software is a global technology and IT services provider headquartered in Vietnam, with more than 33,000 employees operating across 30+ countries. The company specializes in digital transformation, AI, cloud computing, data analytics, software engineering, smart factories, cybersecurity, and industry-specific solutions for sectors including financial services, manufacturing, automotive, healthcare, aviation, and telecommunications. In recent years, FPT Software has accelerated its AI-first strategy, investing heavily in AI-powered platforms, enterprise modernization services, and global partnerships while expanding its footprint across Europe, North America, the Middle East, and Asia-Pacific. Through a combination of large-scale engineering capabilities, consulting expertise, and AI-driven innovation, the company helps enterprises modernize legacy systems, optimize operations, and accelerate digital transformation initiatives worldwide.

You’ve led digital transformation initiatives across aerospace, aviation, enterprise software, and now AI-first delivery platforms at FPT Software. Looking back at your work with Airbus Skywise and large-scale enterprise modernization programs, what lessons directly shaped your vision for Agentic Engineering and Flezi Foundry?

Looking back at Airbus Skywise and large-scale enterprise modernization programs, one key lesson stands out: transformation is rarely constrained by technology. More often, it is constrained by how organizations operate.

With Skywise, we saw the power of combining data, platforms, and ecosystem collaboration. However, even with advanced technology in place, value creation often plateaued when operating models remained human-centric, sequential, and siloed.

That experience directly shaped our vision for Agentic Engineering. We believe the next evolution is not simply better digital platforms, but intelligent, agent-driven delivery systems where AI actively participates in execution rather than serving only as an analytical tool.

Flezi Foundry was built on that principle, helping organizations move from decision-making to AI-augmented delivery models that continuously execute, learn, and improve.

Much of the AI discussion has centered around coding copilots and developer productivity. Why do you believe the more important shift is moving from AI-powered tools to AI-powered operating models?

Coding copilots are important, but they only address a small portion of the value chain. Most of the conversation today focuses on improving developer productivity, while the larger challenge organizations face is fragmented, slow-to-adapt delivery systems.

The real opportunity lies in rethinking how work is structured. AI-powered operating models allow work to be decomposed into tasks that are executed collaboratively by humans and intelligent agents. Agents can participate across planning, execution, validation, and optimization, while feedback loops become continuous and increasingly automated.

At that point, AI stops being just another productivity tool and becomes a system-level capability. That is where organizations begin to realize meaningful enterprise value.

FPT recently launched Flezi Foundry to advance what you describe as AI-augmented delivery for enterprises. What gap in the market were you trying to solve, and how does the platform differ from the growing number of AI development tools already available?

We saw a clear gap in the market. Enterprises have access to an increasing number of AI tools, but many lack a coherent and governed delivery platform that brings those capabilities together.

Most AI solutions focus on isolated activities such as coding assistance or testing automation. They do not address end-to-end delivery orchestration, enterprise governance, or lifecycle integration. As a result, organizations often struggle to scale AI adoption beyond individual use cases.

Flezi Foundry was designed to solve that challenge. It serves as a unified human-agent collaboration platform that embeds AI across the entire software and IT lifecycle while maintaining governance, traceability, and alignment with business outcomes. In that sense, it is not simply another AI tool; it is an operating layer for AI-augmented delivery.

FPT describes Flezi Foundry as a governed human-agent collaboration platform. What does governance actually look like in practice when AI agents are participating in planning, coding, testing, security reviews, and IT operations?

Governance means that every AI action is visible, controlled, and auditable.

In practice, AI agents operate within defined policies that encompass security requirements, compliance obligations, and coding standards. Their actions are logged, traceable, and attributable, while human oversight remains embedded at critical decision points. Data access is managed through established identity and permission controls to ensure that agents only interact with authorized information.

Most importantly, governance is built directly into execution rather than being added afterward. This allows organizations to scale AI adoption while maintaining trust, transparency, and operational control.

One of the biggest concerns enterprises have is accountability. If an AI agent introduces a security vulnerability, deployment issue, or operational failure, how should responsibility be assigned between the AI system, human supervisors, and service providers?

AI does not remove accountability; it redistributes it within a controlled framework.

AI agents execute tasks within defined parameters, but responsibility ultimately remains with the people and organizations that design, supervise, and govern those systems. Human supervisors provide oversight and validation, while service providers are accountable for the architecture, controls, and governance mechanisms that guide agent behavior.

The key is maintaining complete traceability. Every action should be attributable and auditable so that organizations can preserve trust, maintain control, and clearly understand how decisions were made.

Flezi Foundry introduces the concept of an Agentic Development Lifecycle (ADLC). How does this differ from traditional Agile, DevOps, or platform engineering approaches that enterprises already use today?

Agile and DevOps were designed to optimize human collaboration and automate delivery pipelines. The Agentic Development Lifecycle, or ADLC, introduces a fundamentally different element: AI agents as active participants throughout the lifecycle.

Rather than simply supporting developers, agents collaborate with humans to execute tasks, reason through problems, and contribute to decision-making. Continuous reasoning replaces static workflows, while intelligent feedback loops drive ongoing optimization.

A simple way to think about it is that Agile focuses on iterative delivery, DevOps focuses on automated pipelines, and ADLC focuses on adaptive human-agent delivery systems.

Many organizations are experimenting with AI agents, but few have established frameworks for measuring their performance. What metrics do you believe enterprises should use to evaluate AI agents as part of their software delivery workforce?

Organizations should increasingly view AI agents as part of the workforce and measure them accordingly.

Traditional productivity metrics remain important, including delivery velocity (task completion speed) and reductions in manual effort. However, organizations should also evaluate quality indicators such as defect rates, rework levels, accuracy, reliability, and compliance adherence.

Ultimately, the objective is not simply higher productivity. The real measure of success is whether AI agents improve the predictability, quality, and scalability of outcomes.

Flezi Foundry introduces outcome-based pricing models tied to delivery velocity and service-level objectives. Do you see this as the beginning of a broader shift away from traditional FTE-based outsourcing and staff augmentation models?

Yes, we believe a significant shift is already underway.

Traditional FTE-based models were built around the assumption that effort is the primary driver of value. In AI-augmented delivery environments, that assumption becomes increasingly outdated because intelligent agents can significantly change how work is performed.

As a result, outcome-based models tied to delivery velocity, service-level objectives, and measurable business impact represent a more natural fit. Flezi Foundry supports this evolution by making delivery performance more transparent, measurable, and predictable.

As enterprises increasingly deploy AI agents across software development and IT operations, how do you see the role of software engineers evolving over the next five years?

Software engineers will evolve rather than disappear.

As AI assumes a larger share of implementation activities, engineers will increasingly focus on higher-value responsibilities such as system design, architecture, agent orchestration, governance, quality assurance, and business problem framing.

The future role of the engineer will be less about writing every line of code and more about designing systems, guiding intelligent agents, and ensuring that technology outcomes align with business objectives.

Looking ahead, what does a mature human-agent enterprise technology organization look like? If we revisit this conversation in 2030, what aspects of software delivery and IT operations do you think will be fundamentally different from today?

By 2030, I believe leading organizations will operate as true human-agent ecosystems.

AI agents will be embedded throughout delivery and operational processes, enabling continuous execution, real-time optimization, and faster adaptation to changing business needs. Strong governance frameworks will ensure trust, accountability, and compliance, while outcome-based performance models become the standard way organizations measure value.

Software delivery and IT operations will be significantly faster, more adaptive, and more predictable than they are today. The conversation will no longer center on whether organizations should adopt AI agents. Instead, the differentiator will be how effectively they orchestrate humans and agents together at scale.

Thank you for the great interview, readers who wish to learn more should visit FPT Software.