Skill Framework

The AECM Framework

The Augmented Engineer Competency Model defines 8 skill domains that capture what it means to be an effective AI-augmented software engineer. Each domain is independently assessed, tracked, and developed.

8 domains|25 sub-skills|5 proficiency levels

Why a new competency model?

AI tools have changed what it means to be productive as a software engineer. Writing code is no longer the bottleneck — the bottleneck is specifying what to build, verifying the output, and orchestrating the human-AI workflow.

Traditional engineering skill frameworks focus on language proficiency, system design, and algorithmic thinking. These skills remain important, but they no longer tell the full story. AECM fills the gap by defining the skills that determine how effectively an engineer collaborates with AI tools.

The 8 Domains

Each domain represents a distinct category of skills. Together they form a complete picture of augmented engineering competency.

DS

Decomposition & Specification

The ability to break down complex problems into well-specified, AI-executable tasks with clear constraints and acceptance criteria.

DS-1Task Granularity

Breaking work into appropriately-sized units that an AI tool can execute effectively.

DS-2Constraint Articulation

Clearly defining edge cases, error conditions, and non-functional requirements.

DS-3Context Boundary Definition

Determining what context an AI tool needs and scoping information appropriately.

DS-4Acceptance Criteria Specification

Defining measurable success criteria for each task that enable verification.

CO

Context Orchestration

The ability to create and maintain the contextual environment that enables effective AI collaboration.

CO-1Context Provisioning

Providing AI tools with the right information at the right time.

CO-2Documentation Architecture

Structuring project documentation for effective AI consumption.

CO-3Tool Configuration

Setting up AI tool configurations (CLAUDE.md, MCP servers, etc.) for optimal collaboration.

VQ

Verification & Quality Assurance

The ability to critically evaluate AI-generated output for correctness, security, performance, and maintainability.

VQ-1Code Review Acuity

Identifying bugs, logic errors, and quality issues in AI-generated code.

VQ-2Security Assessment

Detecting security vulnerabilities in AI-generated output.

VQ-3Performance Evaluation

Identifying performance issues and optimization opportunities.

VQ-4Architecture Assessment

Evaluating structural quality and maintainability of generated code.

IR

Iterative Refinement

The ability to efficiently guide AI tools toward desired outcomes through feedback and iteration.

IR-1Feedback Precision

Providing clear, actionable feedback that moves AI output toward the target.

IR-2Iteration Efficiency

Achieving desired outcomes in fewer rounds of interaction.

IR-3Approach Pivoting

Recognizing when to change strategy vs. continue refining the current approach.

DT

Domain Translation

The ability to translate business domain knowledge into technical specifications that preserve domain constraints.

DT-1Constraint Preservation

Ensuring domain-specific rules and regulations are maintained in technical output.

DT-2Assumption Detection

Identifying incorrect assumptions an AI tool might make about the domain.

DT-3Domain Specification

Articulating domain requirements in a way AI tools can use effectively.

TE

Tool Ecosystem Management

The ability to select, configure, and orchestrate the right AI tools and workflows for each task.

TE-1Tool Selection

Choosing the appropriate AI tools and configurations for specific tasks.

TE-2Workflow Design

Designing efficient workflows that leverage multiple tools effectively.

AP

Augmentation Prioritization

The ability to determine which tasks benefit from AI augmentation and design the optimal human-AI workflow.

AP-1Task Classification

Accurately classifying tasks as AI-first, manual-first, or hybrid.

AP-2Workflow Optimization

Designing the optimal balance of human and AI effort for each task.

KC

Knowledge Curation

The ability to build and maintain knowledge artifacts that enhance AI collaboration over time.

KC-1Documentation Completeness

Creating comprehensive documentation that enables effective AI assistance.

KC-2Knowledge Evolution

Maintaining and updating knowledge artifacts as projects evolve.

KC-3AI-Ready Documentation

Structuring documentation specifically for AI tool consumption.

Discover your skill profile

Take the free onboarding assessment and see where you stand across all 8 AECM domains.

The Augmented Engineer Competency Model. Assess. Learn. Grow. Prove.