Computer Systems Engineers/Architects

Research, test, or verify proper functioning of software patches and fixes.

AI: Fully automatable - AI can research known fixes, run automated test suites, static/dynamic analysis, and continuous-integration pipelines to test and verify patches at scale with high reliability.

Provide technical guidance or support for the development or troubleshooting of systems.

AI: Fully automatable - AI systems in 2025 can analyze logs/configs, synthesize diagnostics, and provide actionable development and troubleshooting guidance across common scenarios without human mediation.

Document design specifications, installation instructions, and other system-related information.

AI: Fully automatable - AI can reliably produce structured design specifications, installation instructions, and system documentation from specifications, code, and input requirements.

Communicate project information through presentations, technical reports, or white papers.

AI: Fully automatable - AI can synthesize technical content into polished presentations, reports, and white papers tailored to audiences and objectives.

Monitor system operation to detect potential problems.

AI: Fully automatable - AI-driven monitoring and anomaly-detection tools can continuously analyze telemetry and logs to detect potential problems and generate actionable alerts.

Provide customers or installation teams guidelines for implementing secure systems.

AI: Fully automatable - AI can produce tailored, standards-aligned guidelines and checklists for customers and installers to implement secure systems based on best practices and threat models.

Investigate system component suitability for specified purposes and make recommendations regarding component use.

AI: Fully automatable - AI can analyze datasheets, compatibility, performance data, and run simulations to assess component suitability and make defensible recommendations from available evidence.

Complete models and simulations, using manual or automated tools, to analyze or predict system performance under different operating conditions.

AI: Fully automatable - AI and automated tools can set up, execute, and analyze models and simulations across operating conditions to predict system performance with high automation.

Configure servers to meet functional specifications.

AI: Fully automatable - AI can fully automate server configuration by generating and executing infrastructure-as-code and configuration-management scripts when given appropriate access and constraints.

Verify stability, interoperability, portability, security, or scalability of system architecture.

AI: Partial - AI tools can run tests, simulations, and security scans to evaluate stability/interoperability/portability/scalability, but full verification and architectural sign-off still need human assessment.

Develop system engineering, software engineering, system integration, or distributed system architectures.

AI: Partial - AI can generate and iterate architecture proposals and perform tradeoff analysis, but final system engineering and integration design decisions require human expertise and accountability.

Collaborate with engineers or software developers to select appropriate design solutions or ensure the compatibility of system components.

AI: Partial - AI can analyze compatibility and recommend design options, yet real collaboration with engineers to choose and adapt solutions involves negotiation and contextual judgment from humans.

Identify system data, hardware, or software components required to meet user needs.

AI: Partial - AI can identify candidate data, hardware, and software components from requirements and catalogs, but validating choices against organizational constraints and tacit needs needs human validation.

Communicate with staff or clients to understand specific system requirements.

AI: Partial - AI can conduct structured interviews and extract requirements from conversations, but understanding nuanced, conflicting, or political requirements typically requires human-led communication.

Provide advice on project costs, design concepts, or design changes.

AI: Partial - AI can generate design concepts and rough cost estimates and suggest changes, but lacks full contextual knowledge, market-specific pricing accuracy, and accountability to fully own project cost/design decisions.

Perform security analyses of developed or packaged software components.

AI: Partial - AI can perform automated static analysis, vulnerability scanning, and code-review guidance, but cannot fully replace human-led, context-aware security assessments and exploit validation.

Define and analyze objectives, scope, issues, or organizational impact of information systems.

AI: Partial - AI can define objectives, scope, and analyze impacts using frameworks and available data, but cannot fully capture organizational politics, tacit knowledge, or lead stakeholder negotiations alone.

Design and conduct hardware or software tests.

AI: Partial - AI can design and execute software tests (unit/integration/automated) effectively, but hardware test design and physical test execution still require human/physical involvement.

Evaluate current or emerging technologies to consider factors such as cost, portability, compatibility, or usability.

AI: Partial - AI can research and compare technologies on cost, compatibility, portability, and usability, but cannot fully validate real-world performance or organizational fit without empirical pilots.

Establish functional or system standards to address operational requirements, quality requirements, and design constraints.

AI: Partial - AI can draft standards from requirements and best practices and check consistency, but cannot fully establish, enforce, and negotiate standards without human governance and stakeholder approval.

Develop or approve project plans, schedules, or budgets.

AI: Partial - AI can generate detailed project plans, schedules, and budgets, but final approval, resource negotiation, and accountability require human decision-makers.

Develop efficient and effective system controllers.

AI: Partial - AI can design and tune system controllers algorithmically and produce implementable designs, but ensuring safety, edge-case robustness, and regulatory compliance needs human oversight.

Evaluate existing systems to determine effectiveness and suggest changes to meet organizational requirements.

AI: Partial - AI can evaluate system telemetry, run diagnostics, and propose changes, but mapping those recommendations to organizational strategy and trade-offs requires human judgment.

Develop application-specific software.

AI: Partial - AI can produce large portions of application-specific software and prototypes, but complex requirements gathering, integration, and acceptance testing still need significant human involvement.

Perform ongoing hardware and software maintenance operations, including installing or upgrading hardware or software.

AI: Partial - Software maintenance (patching, upgrades) is largely automatable, but hardware installation and physical maintenance steps require human intervention, so the overall task is only partially automatable.

Direct the installation of operating systems, network or application software, or computer or network hardware.

AI: Partial - AI can fully automate OS/software installs and orchestrate deployments via scripts, but cannot perform physical hardware installation and often requires human oversight for complex environments.

Train system users in system operation or maintenance.

AI: Partial - AI can generate and deliver interactive training materials and assessments, but hands-on guidance, context-specific mentoring, and organizational buy-in typically require human trainers.

Direct the analysis, development, and operation of complete computer systems.

AI: Not automatable - AI cannot assume the managerial responsibility, legal accountability, and human leadership required to direct analysis, development, and operation of entire systems.