Developer Competency Framework: Dreyfus Model for Performance Reviews

The Dreyfus Model of Skill Acquisition

Five stages backed by cognitive science research, not HR buzzwords

Novice → Advanced Beginner → Competent → Proficient → Expert

Rigid rules Situational patterns Mental models Holistic perspective Intuitive mastery

Why Dreyfus for Developers?

Developed by Stuart and Hubert Dreyfus at UC Berkeley (1980s) through Air Force-funded research on pilot training. Engineers recognize themselves in these stages because the model describes actual cognitive patterns in skill development, not arbitrary corporate levels.

The model maps naturally to career ladders: Junior → Mid-Level → Senior → Staff → Principal. It acknowledges that experts think differently than beginners, not just "better."

Core Insight: Beginners rely on explicit rules. Experts rely on intuition and pattern recognition. This isn't just "knowing more"—it's a fundamental shift in how the brain processes expertise. See research →

But technical skill is only half the picture. An Expert developer who waits to be told what to do creates more management overhead than a Competent developer with high agency. See how Dreyfus combines with ownership mindset in the Performance Matrix →

Visual representation of the five Dreyfus stages, showing the progression from rigid rule following to intuitive expertise

1. Novice

Thinking style: Follows rules rigidly. Needs step-by-step instructions.

Career level: Intern, Junior Developer

"Like a computer following a program" — Dreyfus, 1986

In practice: Writes code from detailed tickets. Struggles with ambiguous requirements. Needs explicit technical specs.

2. Advanced Beginner

Thinking style: Recognizes recurring patterns from experience. Applies situational judgment.

Career level: Junior to Mid-Level Developer

"Learns to recognize meaningful situational aspects"

In practice: Debugs familiar problems. Applies known solutions to similar situations. Struggles with novel challenges.

3. Competent

Thinking style: Develops mental models. Prioritizes relevant features. Takes emotional responsibility.

Career level: Mid-Level to Senior Developer

"Chooses a plan that organizes the situation"

In practice: Designs features independently. Makes architectural decisions for bounded contexts. Weighs trade-offs.

4. Proficient

Thinking style: Sees situations holistically via pattern recognition. Intuition guides what to do.

Career level: Senior to Staff Engineer

"Immediately struck by the appropriate action"

In practice: Anticipates problems before they manifest. Mentors effectively. Sees system-level implications of local changes.

5. Expert

Thinking style: Both perception and response become intuitive. Acts "by feel" without calculating.

Career level: Staff, Principal, Distinguished Engineer

"Sees intuitively what to do without applying rules"

In practice: Designs novel architectures. Identifies elegant solutions to complex problems. Influences industry practices.

Critical Insight: Domain-Specific Expertise

An engineer can be an Expert in one domain (e.g., distributed systems) while being Competent or Advanced Beginner in another (e.g., frontend frameworks). The Dreyfus model applies per skill area, not to "the whole person."

Technical Skill Is Only Half the Picture

The Dreyfus Model measures what people know. But performance also depends on how they work—their ownership mindset and agency level.

See the Performance Matrix →

Blik's Software Engineering Competency Framework

10 dimensions adapted from research and industry best practices

Framework Inspiration

Our default questionnaire synthesizes insights from:

Dreyfus Model of skill acquisition (see above)
Open-source 360-degree engineering self-review frameworks
Industry-standard competency models from high-performing engineering teams

1. Technical Proficiency

What it measures: Code quality, problem-solving ability, and technical depth.

Novice creates code from well-defined requests. Expert conquers complex problems with elegant solutions.

2. Personality & Collaboration

What it measures: Teamwork, interpersonal skills, cultural fit.

Would colleagues actively choose to work with this person?

3. Technical Breadth

What it measures: Technology awareness, cross-platform capability, learning agility.

From narrow specialization to broad experience across stacks and paradigms.

4. Intrinsic Motivation

What it measures: Genuine passion for programming beyond work requirements.

Does this person code for the love of it, or just for paychecks?

5. Engineering Maturity

What it measures: Application of best practices: testing, design patterns, agile methodologies.

From ad-hoc coding to systematic engineering discipline.

6. Pragmatism

What it measures: Balancing technical ideals with business realities and constraints.

Shipping practical solutions vs. pursuing theoretical perfection.

7. Team Cooperation

What it measures: Mentoring, code review quality, supporting team objectives.

Individual contributor vs. force multiplier for the team.

8. Communication

What it measures: Explaining technical concepts, justifying design decisions, documentation.

Can they make complex ideas understandable to varied audiences?

9. Growth Potential

What it measures: Learning capacity, adaptability, receptiveness to feedback.

Trajectory matters as much as current state.

10. Strategic Vision

What it measures: Product thinking, architectural foresight, anticipating future challenges.

Tactical execution vs. strategic thinking about system evolution.

Customizable for Your Context

This is the default framework. Blik is open source—modify dimensions, adjust weighting, add domain-specific competencies. The framework serves your needs, not the other way around.

Examples: Add "Security Awareness" for fintech teams. Include "Research Ability" for ML teams. Remove dimensions that don't apply to your context.

See the Framework in Action

360 Feedback Report with Dreyfus Framework - Light Theme

360 Feedback Report with Dreyfus Framework - Dark Theme

Hover to toggle theme

Radar charts visualize competency ratings across all 10 dimensions, with breakdowns by reviewer category

Why This Approach Works for Engineering Teams

Research-Based

Dreyfus model has 40+ years of cognitive science research. Not invented by consultants, but by academics studying actual skill acquisition.

Engineers Recognize Themselves

The stages map to lived experience. Developers understand where they are and what growth looks like without translation from HR jargon.

Domain-Specific Application

Acknowledges that someone can be Expert in databases but Novice in machine learning. Avoids oversimplified "levels."

Culture and Skills

Combines technical evaluation (Dreyfus) with cultural fit (Sunday Test). Both matter for high-performing teams.

Actionable Feedback

Clear progression from one stage to the next. Reviewees understand what capabilities to develop for advancement.

Open and Adaptable

Framework is a starting point, not dogma. Fork it, customize it, improve it. Share improvements with the community.

Example: Evaluating "Code Quality"

How Dreyfus stages apply to a specific competency

Stage	Code Quality Characteristics
Novice	Code works for the happy path. Inconsistent formatting. Hard-coded values. Minimal error handling.
Advanced Beginner	Follows style guide. Extracts some magic numbers to constants. Adds basic error handling. Code is readable.
Competent	Well-structured modules. Appropriate abstractions. Comprehensive error handling. Tests cover main scenarios. Considers edge cases.
Proficient	Clear separation of concerns. Anticipates future changes. Excellent test coverage. Code is self-documenting. Patterns applied appropriately, not dogmatically.
Expert	Elegant solutions to complex problems. Code that other experts reference. Creates reusable components. Deep understanding of performance and maintainability trade-offs.

Research Foundation & Sources

Over 40 years of peer-reviewed cognitive science research

Academic Credibility

40+

Years of research

1980

Original publication

10+

Professional domains

Commissioned by U.S. Air Force. Validated across nursing, medicine, software engineering, aviation, and education.

Primary Research (Original Dreyfus Work)

Studied skill acquisition in pilots, chess players, and other expert performers. Focus: How humans progress from rule-based thinking to intuitive expertise.

Dreyfus, S. E., & Dreyfus, H. L. (1980). A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition. University of California, Berkeley, Operations Research Center.

The foundational 18-page report. Freely available.

Read PDF (DTIC) Google Scholar

Dreyfus, H. L., & Dreyfus, S. E. (1986). Mind Over Machine: The Power of Human Intuition and Expertise in the Era of the Computer. Free Press.

The definitive book elaborating the model. Most-cited reference.

Goodreads Google Scholar

Dreyfus, S. E. (2004). The Five-Stage Model of Adult Skill Acquisition. Bulletin of Science, Technology & Society, 24(3), 177-181.

Concise academic summary of the model.

DOI Link Google Scholar

Core Insight

The Dreyfus brothers discovered that experts don't just know more—they think fundamentally differently. Experts rely on intuition and pattern recognition developed through extensive experience, while novices depend on explicit rules and procedures.

Four Mental Functions

The model tracks progression through four dimensions: recollection (what you remember), recognition (what patterns you see), decision (how you choose), and awareness (how you perceive situations).

Cross-Domain Validation

The model has been successfully applied in aviation, chess, nursing, medicine, and software engineering—proving it describes universal patterns of how humans acquire expertise, not domain-specific quirks.

Applications in Professional Domains

Software Engineering

Hunt, A. (2008). Pragmatic Thinking and Learning: Refactor Your Wetware. Pragmatic Bookshelf.

Andy Hunt (Pragmatic Programmer co-author, Agile Manifesto signatory) applied Dreyfus to software development teams.

Publisher Site Goodreads

Nursing & Healthcare

Benner, P. (1984). From Novice to Expert: Excellence and Power in Clinical Nursing Practice. Addison-Wesley.

Most influential application - became foundational to nursing education worldwide.

Google Scholar Goodreads

Comprehensive Multi-Domain Analysis

Mangiante, E. S., & Peno, K. (Eds.). (2021). Teaching and Learning for Adult Skill Acquisition: Applying the Dreyfus and Dreyfus Model in Different Fields. Information Age Publishing.

Applications across healthcare, education, law enforcement, business, military, and ethics training.

Google Scholar

Why This Research Matters for Engineering Teams

Understanding the Dreyfus stages prevents common management mistakes:

Don't expect novices to make expert decisions: Junior engineers need explicit rules and step-by-step instructions. Asking them to make architectural decisions sets them up for failure.
Don't constrain experts with rigid checklists: Senior engineers rely on pattern recognition and intuition. Forcing them to follow novice-level procedures wastes their expertise.
Recognize domain-specific expertise: An engineer can be Expert in distributed systems while being Novice in machine learning. This isn't inconsistency—it's how skill acquisition actually works.

The Dreyfus model gives you a research-backed vocabulary for discussing skill development that engineers actually understand and recognize in themselves.

Competency Frameworks That Engineers Actually Respect

TL;DR