The landscape of software engineering is undergoing a profound transformation, a shift so significant that it’s redefining the very essence of the individual contributor role. Cameron Etezadi, CTO at LaunchDarkly and a former VP of Engineering at IBM, posits that the advent of Artificial Intelligence (AI) has effectively elevated every software engineer, regardless of experience level, into a de facto front-line manager. This assertion, shared in a recent discussion with The New Stack, suggests that the skillset now paramount for engineers mirrors that of effective managers: robust project planning and adept cross-team coordination. Etezadi advocates for a recalibration of how engineering performance is evaluated, urging companies to adopt frameworks typically reserved for assessing front-line managerial effectiveness.
This perspective gains traction when examined alongside emerging research from Gartner. The analyst firm projects a significant trend towards consolidation within software engineering teams, predicting that by 2029, 60% of organizations will operate with smaller engineering units. Aliyah Camacho, a Principal Analyst at Gartner, anticipates that AI will be a primary catalyst for the rise of "tiny teams," potentially comprising as few as two to three engineers. If the traditional software engineering role is indeed poised to shrink or evolve into a middle-management-adjacent position, it raises critical questions about whether AI is truly enhancing engineer productivity or merely altering its manifestation.
The Elusive Definition of Engineering Productivity in the AI Era
The pervasive integration of AI coding tools, heralded by major technology firms as the unequivocal future of software development, prompts a crucial debate: does the ability to generate more code, faster, equate to genuine productivity gains? Daniel Wang, CTO of Citizen Health and formerly Director of Engineering at Uber, challenges this direct correlation. "Software exists to solve problems," Wang stated to The New Stack. "If AI lets us ship 10x as much code but customer outcomes don’t improve, we aren’t more productive."
Wang emphasizes a critical paradigm shift in how productivity should be measured. He argues that the focus must move beyond mere code output to encompass the quality of decisions and the ultimate impact on customer outcomes. This necessitates a comprehensive evaluation of the entire development lifecycle, from the initial ideation phase through to deployment and beyond. Key metrics, according to Wang, should include cycle time from concept to production, rollback rates, the incidence of escaped defects, and overall system reliability. He also champions the incorporation of qualitative assessments, posing questions such as, "Did we choose the right solution?" and "Does the code actually solve our customers’ problems?" These fundamental inquiries, while seemingly self-evident, are often overlooked in current assessments of AI’s impact on engineering efficiency.
This disconnect is further illuminated by insights from Ameya Kanitkar, founder and CTO of AI measurement platform Larridin and former Head of Engineering at Coinbase. Kanitkar observes that many companies continue to rely on traditional, often superficial, metrics. "It’s largely tangible but irrelevant activity," he noted, referring to metrics such as lines of code, pull requests merged, or velocity points – the same indicators used before the widespread adoption of AI. Kanitkar contends that these metrics fail to provide a genuine gauge of value and, in fact, can be actively detrimental. "Living and dying by these metrics could actually steer engineering teams away from truly productive customer outcomes," he warned. If teams are incentivized by code volume and commit frequency – metrics easily inflated by AI – they may inadvertently prioritize quantity over substantive value creation.
The Paradox of Productivity and Burnout
The pressure to deliver increased output at an accelerated pace, amplified by the capabilities of AI, is creating a paradoxical situation: engineers report feeling more productive yet simultaneously experiencing profound exhaustion. Kanitkar has observed a growing trend where engineers utilize AI agents to perform tasks overnight, a practice that, while seemingly efficient, fosters a continuous underlying pressure. "We also see engineers wanting to set agents loose overnight so the work is waiting for them in the morning, which sounds efficient, but it creates this constant background pressure to keep feeding the agents new work so they’re never sitting idle. That pressure is draining," he explained.
This sentiment echoes a widely shared observation by David Holz, founder of Midjourney, who recently posted on X: "my friends are all feeling extremely productive and also extremely drained with the latest coding models. this makes me feel like something is wrong, and also that there might be a big opportunity. does anyone have any strategies they use to make it feel better day-to-day?" This widespread feeling suggests a fundamental disconnect between perceived output and actual well-being, a chasm exacerbated by the scarcity of reliable methods for companies to measure AI’s true impact on engineering team performance and outcomes.
Kanitkar attributes this fatigue directly to the evolving managerial aspects of the engineer’s role. The constant need to context-switch and manage multiple AI agents simultaneously, he suggests, is a significant contributor to burnout. As Gartner’s projections indicate a future of smaller engineering teams, this intensive "agent-babysitting" could become the new standard. However, the ultimate question of whether this AI-augmented, manager-like engineering paradigm leads to genuinely superior productivity remains an open and critical inquiry.
The Shifting Skillset: From Code to Coordination
The assertion that AI is transforming individual contributors into managers is not merely an observation of changing workflows; it implies a fundamental redefinition of essential engineering skills. Traditionally, the emphasis for individual contributors has been on deep technical proficiency, problem-solving through code, and algorithm optimization. While these skills remain vital, Etezadi’s argument suggests that their relative importance is diminishing in favor of a broader, more strategic skillset.
The ability to decompose complex problems into manageable tasks, effectively plan project timelines, allocate resources (including AI agents), and communicate technical strategies across diverse teams are becoming hallmarks of successful engineers. This managerial aptitude extends to understanding project scope, anticipating potential roadblocks, and ensuring alignment with broader business objectives – tasks previously relegated to project managers or senior leadership. The "fresh out of school" engineer, in this new paradigm, is expected to possess not only coding prowess but also the organizational and communication skills akin to an entry-level team lead.
This evolution can be traced back to the increasing complexity of modern software systems and the accelerated pace of technological advancement. AI tools, while powerful, are not autonomous problem-solvers in the strategic sense. They require precise direction, iterative refinement, and careful integration into larger workflows. Engineers are increasingly tasked with framing the problems for AI, evaluating its outputs critically, and ensuring that the generated code or solutions align with architectural principles and business requirements. This requires a higher level of abstract thinking and strategic oversight, functions typically associated with management.
The Managerial Metamorphosis: Implications for Talent Development and Evaluation
If Etezadi’s assessment holds true, organizations must fundamentally rethink their approach to talent development and performance evaluation within engineering departments. Traditional performance reviews that solely focus on lines of code or feature delivery may become obsolete. Instead, evaluations should incorporate criteria such as:
- Project Planning and Execution: The ability to break down complex projects, set realistic timelines, and manage dependencies.
- Cross-Team Collaboration: Effectiveness in communicating with product managers, designers, QA, and other engineering teams to ensure alignment and smooth integration.
- Problem Decomposition: Skill in clearly defining problems for AI tools and human team members alike.
- Outcome-Oriented Thinking: A demonstrated focus on delivering tangible business value and improving customer experiences, rather than simply producing code.
- Risk Management and Mitigation: Proactive identification of potential issues and development of strategies to address them.
The implications for engineering education are also significant. Universities and bootcamps may need to integrate more coursework focused on project management, systems thinking, and effective communication alongside traditional computer science curricula. The development of "soft skills" will become as critical as mastering programming languages and data structures.
The rise of "tiny teams" as predicted by Gartner further underscores this managerial shift. With fewer individuals on a team, each engineer will likely bear a greater responsibility for the end-to-end success of their projects. This necessitates a high degree of autonomy, self-management, and the ability to coordinate efforts, even if those efforts involve collaborating with AI agents rather than solely human colleagues.
The Future of Engineering: A Collaborative Ecosystem
The future of software engineering appears to be a dynamic ecosystem where human ingenuity and artificial intelligence collaborate. While AI tools can automate repetitive tasks and accelerate code generation, the strategic direction, critical evaluation, and ultimate problem-solving remain firmly in the human domain. The engineers of tomorrow will likely be less concerned with writing every line of code and more focused on orchestrating complex systems, guiding AI tools effectively, and ensuring that technological innovation directly translates into meaningful business and customer value.
The transition is not without its challenges. The potential for burnout, the difficulty in accurately measuring true productivity, and the need for a significant shift in organizational evaluation metrics are all critical considerations. However, by embracing the evolving nature of the software engineer’s role and equipping them with the necessary skills and frameworks, the industry can navigate this transformation successfully, unlocking new levels of innovation and efficiency. The era of the purely individual contributor may be drawing to a close, replaced by a more multifaceted, managerially inclined, and AI-augmented software engineer.
