AI is making engineering teams smaller and roles bigger

AI expands the software engineer’s job

AI is changing what companies pay engineers to do. Writing code remains important, but code production is no longer the only constraint. As AI agents generate more software, the harder problem becomes deciding whether that output is correct, secure and fit for the business purpose.

A May report from software platform Harness found that AI adoption in engineering workflows has become the default. This changes the engineer’s scope of responsibility. Engineers must review AI-generated code, test its quality and security, and decide when to accept or override an AI agent’s work. They also remain accountable for downstream outcomes. AI can produce code, but it cannot remove organizational accountability for what that code does in production.

This distinction matters for executives. Faster code generation has limited value if review, testing and security processes cannot keep pace. AI can shift work away from implementation and toward verification. Companies therefore need to measure the full software delivery process.

The skill profile must change with the work. Engineers need stronger judgment in areas such as system design, security, testing and product requirements. They also need to understand how AI agents behave, where their output can fail and when human intervention is necessary. This does not make core engineering knowledge less important. It makes that knowledge more important because engineers must assess work they did not necessarily produce themselves.

For leadership teams, the priority is governance and capability. Define who owns AI-generated code. Set clear standards for review and security. Give engineering teams the tools and training needed to validate automated output. Companies that automate code creation without strengthening these controls risk increasing output faster than their ability to assure its quality.

The near-term opportunity is clear. AI can reduce time spent on routine implementation and move engineering capacity toward higher-value decisions. But productivity should be judged by reliable software delivered against business goals. The organizations that make this shift successfully will treat AI as a change in engineering responsibility.

Smaller teams become the new engineering model

AI is reducing the number of separate roles needed to move software from an idea to a working product. Gartner’s report suggests organizations may get better results by organizing employees into smaller, multidisciplinary teams as traditional engineering responsibilities change. The objective is to reduce coordination overhead while giving each team enough capability to deliver an outcome.

Gartner describes a potential team that includes a product manager, a user experience or agent experience designer, and at least one AI-native software engineer. An AI-native engineer is comfortable using AI agents as part of everyday development rather than treating AI as a separate tool. Across the team, responsibilities can include understanding business goals, designing the product and directing or reviewing work performed by AI agents.

This structure changes the management problem. When AI can handle more implementation work, adding people does not automatically increase output. Coordination, decision speed and quality control can become the limiting factors. Smaller teams can reduce handoffs between product, design and engineering. They can also give individuals clearer ownership of the result rather than ownership of a narrow task.

Team size still matters. Gartner does not prescribe one fixed number of employees for every project. Camacho, in connection with Gartner, said teams should remain nimble while being large enough to support a diversity of ideas. The appropriate structure therefore depends on the product, technical complexity and objectives. A small team building a contained feature will have different requirements from one responsible for a critical enterprise system.

Executives should also avoid interpreting multidisciplinary work as evidence that specialist expertise is no longer necessary. AI may broaden what each employee can accomplish, but security, architecture, user experience and other high-risk disciplines can still require deep expertise. The practical goal is to remove unnecessary organizational boundaries without removing controls or skills that protect product quality.

The operating model should follow the work. Leaders should identify where handoffs slow delivery, determine which tasks AI can perform reliably, and assign explicit human accountability for important decisions. They should then design teams around business outcomes rather than existing job boundaries.

The result can be a leaner and faster engineering organization. But smaller teams are useful only when AI genuinely increases individual capacity and employees have the skills to manage that broader scope. Team compression should therefore be treated as an organizational design decision.

AI pushes engineers toward product ownership

AI is changing the boundary between software engineering and product management. As AI takes on more implementation work, engineers can spend more time deciding what to build, how it should work and whether the result meets the business need. Gartner describes this shift as software engineers becoming “product engineers.”

Camacho, in connection with Gartner, said: “Due to AI-driven compression of roles and competencies, software engineers are becoming ‘product engineers,’ freeing product managers to focus on the product vision and feature roadmap of the future rather than feature implementation details.”

The important change is ownership. An engineer working as a product engineer needs more than coding ability. The role requires an understanding of customer needs, business goals and product design, alongside the technical judgment needed to manage AI agents and validate their output. Engineers therefore become more involved in decisions that were previously distributed across several functions.

Product managers also gain a clearer mandate. If engineers can take greater responsibility for implementation details and product-level execution, product managers can focus more attention on future priorities. That includes defining product direction, deciding which problems deserve investment and maintaining a roadmap that supports business objectives. AI creates value here by changing how human time is allocated.

For executives, role clarity becomes critical. Broader responsibilities can improve decision speed, but unclear ownership can create the opposite result. Leaders need to specify who makes product decisions, who approves technical decisions and who remains accountable for AI-generated work. Expanding an engineer’s scope should not mean duplicating the product manager’s authority.

This also changes hiring and development priorities. Technical depth remains essential, but engineers who can connect technical choices to customer and commercial outcomes become more valuable. Companies should develop these capabilities deliberately rather than assuming that access to AI tools will create product judgment.

The strategic benefit is a tighter connection between product intent and technical execution. Product managers can spend more time planning what comes next, while engineers take broader responsibility for turning those priorities into reliable products. Companies that define these roles clearly can use AI to reduce implementation friction without weakening technical or product accountability.

Cutting junior hiring creates a long-term talent risk

AI can reduce the amount of routine work assigned to junior software engineers. That creates an immediate temptation for companies to hire fewer entry-level employees. Gartner warns that this response can create a larger problem later. Companies still need a reliable way to develop the senior engineers they will depend on in future.

Camacho, in connection with Gartner, stressed that Gartner is not predicting that AI-driven restructuring should result in job losses. Instead, the concern is how companies redesign their workforce as roles change. Slower junior hiring may reduce costs in the short term, but it can also weaken knowledge transfer, disrupt internal talent pipelines and leave companies more dependent on expensive senior hires.

This matters because senior engineering capability takes time to develop. Experience with production systems, architecture, security and business requirements comes through sustained work and increasing responsibility. AI can help junior engineers perform tasks and acquire information faster, but access to AI does not automatically produce the judgment needed to make high-impact technical decisions.

The economics can also shift in an unexpected direction. If many companies reduce junior recruitment at the same time, fewer engineers will gain the experience required for senior positions. Employers could then face stronger competition for an already valuable group of experienced professionals. Gartner’s argument is therefore not simply about preserving entry-level jobs. It is about maintaining the supply and development of critical technical skills.

For executives, the key decision is how much junior work to automate without weakening workforce development. Entry-level roles will likely need to change. Junior engineers can spend less time on basic code production and more time learning how to validate AI output, test systems, understand product requirements and work with experienced engineers. That approach uses AI to improve development rather than eliminating the development path itself.

Camacho said, “AI is reshaping software engineering, not by replacing developers, but by creating a surge in demand for intelligent applications and new engineering roles.” The supplied text does not provide Camacho’s first name or formal Gartner position, so those details should not be inferred.

Leadership teams should therefore treat workforce planning and AI adoption as connected decisions. Productivity gains today should not come at the expense of the skills the organization will need in three, five or ten years. Companies that continue recruiting and developing early-career engineers while redesigning their work around AI will be better positioned to maintain technical capability as engineering roles evolve.

Key takeaways for decision-makers

• AI shifts engineering toward oversight: AI can generate more code, but engineers remain responsible for quality, security and downstream outcomes. Leaders should strengthen review standards, governance and technical skills as automation expands.
• Smaller teams require broader capabilities: AI can reduce coordination needs and let multidisciplinary teams handle more work with fewer handoffs. Design teams around business outcomes and required expertise rather than using AI as a simple headcount reduction tool.
• Engineers are gaining product ownership: AI is moving engineers beyond implementation toward product decisions, while product managers can focus more on vision and roadmaps. Executives should clarify decision rights and develop engineers who can connect technical choices with customer and business needs.
• Protect the junior talent pipeline: Cutting entry-level hiring may lower near-term costs but weaken the future supply of experienced engineers. Redesign junior roles around AI-assisted development, validation and learning rather than removing the pathway to senior expertise.