AI's Rise Renews Focus on Forward-Deployed Engineers

The current focus on this role prompts a critical examination to determine if it truly offers a proven model for navigating the complexities of AI development and deployment, or if it is merely an industry myth.

This session aims to leave attendees with a working definition of the role, a comprehensive competency model, and a practical playbook along with a different lens through which to view its application.

Harinee Muralinath serves as the Director and Business Information Security Officer at ThoughtWorks, where she oversees security initiatives across India and the Middle East, as well as the firm's AI projects. Her career is uniquely positioned at the intersection of security, quality engineering, and AI delivery.

This background means she has spent years observing organizations repeatedly address the same underlying problems under various names, providing her with a comprehensive perspective on evolving engineering roles and challenges.

It also seeks to establish a comprehensive competency model that covers the entire scope of the job.

Attendees will be equipped with a practical playbook and a fresh perspective, or "lens," for effectively applying this role within their organizations.

Clients, whether external in a services industry or internal within a product organization, do not initiate engagements by requesting specific roles like a "forward-deployed engineer." Instead, they articulate business needs driven by factors such as market competition, regulatory pressures, or demanding boards.

Every client operates under significant resource constraints, primarily concerning money, time, and attention, which dictates their focus on problem resolution rather than role titles. Ultimately, every technical role, including that of a forward-deployed engineer, exists solely to produce a specific, measurable outcome that addresses these core business needs.

Many industry discussions and presentations about the Forward-Deployed Engineer (FDE) role often fail to provide a precise and formal definition. There is a common assumption within the industry that everyone implicitly understands what the role entails, leading to a lack of explicit articulation.

Consequently, job descriptions across the industry exhibit significant inconsistency in how they define the FDE role and its core responsibilities. This ambiguity can hinder effective hiring, training, and integration of FDEs into teams, despite their growing importance in the AI landscape.

Historically, the costs associated with software development were high, primarily driven by the labor-intensive aspects of execution, coding, and debugging. Generative AI (GenAI) has significantly collapsed these execution costs, enabling engineers to shift their focus upstream towards more strategic activities.

However, the cost of judgment—the ability to discern quality, understand context, and know when to escalate issues—has remained largely unchanged. AI does not automatically imbue engineers with critical skills such as discovery, effective stakeholder management, or comprehensive threat modeling, highlighting the enduring value of human expertise in these areas.

Engineers can conceptualize different roles, such as business analysts, testers, and security experts, as distinct "bundles of judgment." A forward-deployed engineer does not aim to replace these specialized roles but rather to learn and adopt their unique "lenses" to understand what each discipline prioritizes and looks for within a system.

For instance, business analysts excel at identifying unspoken requirements, testers question underlying assumptions, and security professionals focus on defining and defending trust boundaries. This cultivated ability to adopt and apply multiple disciplinary perspectives represents a critical cross-functional skill, particularly vital in the evolving age of AI.

By integrating these different lenses, an FDE can anticipate potential issues, facilitate better communication across teams, and ensure a more holistic approach to product development and problem-solving, without needing to become an expert in every single domain.

While AI offers extraordinary breadth by efficiently answering questions within known parameters, it does not automatically provide depth; this must be actively cultivated through deliberate questioning. In the context of security, AI can effectively cover standard risks, such as those identified by OWASP, but it often fails to address the deeper, "under-the-waterline" questions that are crucial for comprehensive protection.

These vital, depth-oriented questions include understanding how trust boundaries might change, identifying potential data leaks within logs or prompts, and establishing robust incident evidence gathering procedures. AI serves as a powerful leverage tool for efficiently answering questions, but the human element remains paramount in meticulously defining what questions to ask to achieve true depth and robustness in solutions.

Current academic white papers and training programs for Forward-Deployed (FD) roles primarily emphasize technical craft and programming skills. While these are foundational, they represent only one aspect of the comprehensive skillset required.

Increasingly, more effective training models are beginning to integrate customer interaction skills as a core competency. The precise definition of these evolving skills, crucial for navigating the complexities of client engagement and problem-solving, continues to adapt as the industry responds to the demands of an AI-driven landscape.

Analysis of forward-deployed engagements reveals that project failures in the field are rarely attributable to issues with the code itself. Instead, common root causes include poorly defined scope that was never properly segmented, and misaligned expectations among various stakeholders who were not adequately considered.

Delayed escalations also significantly contribute to project failures, preventing timely interventions when problems arise. While robust system architecture and quality engineering are undoubtedly vital components of success, they alone are insufficient to guarantee a project's triumph without addressing these human and organizational factors.

Years of development experience do not automatically confer critical skills such as effective discovery, precise problem framing, accurate scoping, or sequential planning under real-world constraints. Similarly, robust risk and dependency management are essential for achieving project predictability.

Clients tend to be more forgiving of projects that progress slowly but consistently, rather than those that deliver unexpected surprises. Furthermore, the ability to influence others without formal authority is an often underrated yet crucial skill for engineers operating within complex client organizations, highlighting the human and management dimensions beyond pure technical delivery.

The first level is "awareness," signifying a basic understanding of a concept or domain.

The second level is "working capability," which means the engineer can safely handle a normal case or routine task independently. The third and highest level is "depth," where the engineer is recognized as an expert whom others would call upon for counsel and complex problem-solving.

They should also adopt another discipline’s "lens" for a month, such as UX or security, to broaden their perspective.

Building one area of real technical depth is crucial, using AI as a sparring partner rather than an oracle to challenge assumptions and deepen understanding. Furthermore, engineers must practice direct client communication and expectation setting, mastering the art of influence without authority by building an escalation map to proactively manage risks and dependencies.

Knowing when to seek help is a sign of strength and essential for successful project delivery.

Integrating client teams into the delivery process is critical for fostering transparency and shared ownership. Ultimately, the focus remains on consistent value delivery, achieved through collective effort and effective teamwork rather than individual heroics.

The individual growth and development of an aspiring Forward-Deployed (FD) engineer are significantly constrained without the presence of a supportive organizational ecosystem. Leadership plays a pivotal role in creating this environment.

Leaders must actively support and invest in the development of forward-deployed capabilities within their teams. This includes providing opportunities for cross-functional learning, fostering a culture of collaboration, and recognizing the unique contributions of FDEs, thereby enabling them to thrive and effectively bridge the gap between technical execution and business outcomes.

Experience, particularly the nuanced judgment required for advanced engineering roles, cannot be treated as an "instant coffee" product; it must be cultivated over time. Leaders have a responsibility to intentionally create environments that foster cross-disciplinary judgment, providing opportunities for engineers to gain diverse exposure.

The successful implementation and growth of the Forward-Deployed Engineer (FDE) model are contingent upon these specific conditions, emphasizing that true expertise is built through structured and deliberate experiential learning, not through shortcuts.

Organizations face two extremes in the age of AI: viewing engineers as superhumans capable of everything or rigidly adhering to outdated, siloed structures. The true opportunity lies in recomposing delivery processes around desired outcomes, rather than falling into these traps.

This involves automating work that does not directly alter the outcome, while simultaneously focusing on growing people to safely manage the wider span of responsibilities that emerge. The goal is to retain and cultivate the human expertise that directly influences the outcome, optimizing for efficiency while nurturing essential human skills.

The subsequent focus should be on identifying the most cost-effective and trustworthy methods to facilitate that critical thinking.

The choice between deploying a Forward-Deployed Engineer (FDE) or forming a cross-functional team should be based squarely on the desired outcome and the most efficient way to achieve it, rather than being swayed by industry hype or an automatic assumption that AI will replace all human roles.