Experience and judgment.
Experts define context, objectives and practical conditions. They question assumptions, assess proposals and retain responsibility for decisions.
ON ORGANIZATIONAL LEARNING
My work explores how people and AI solutions
can learn through organizational data and shared methods.
01 — FROM EXPERIENCE TO LEARNINGMy work explores how organizations can use their own data, defined calculations and business rules to produce reproducible results, and turn those results into lasting problem-solving and decision experience.
My focus is on how people and AI solutions can contribute to organizational learning together by using shared methods, data-backed results and past decision context. I work on structures where calculations are deterministic, and interpretations and proposals can be assessed alongside their supporting evidence.
Alongside “What did we learn?”, I ask:
“What difference did our learning make?”
The connections explored here reflect my practice perspective, informed by the cited sources. I do not present these four approaches as a single model.
MY FOCUS / PEOPLE & AI
How can contributions from people and AI solutions become a shared learning process that an organization can revisit, question and build upon?
I explore how an organization’s calculations, methods and business rules can evolve together with the problem-solving and decision experience built upon them, and how this accumulated experience can remain with the organization as its people or AI solutions change.
Experts define context, objectives and practical conditions. They question assumptions, assess proposals and retain responsibility for decisions.
Within their access permissions, AI tools and agents can use defined methods, deterministic calculation results and past context. They can develop interpretations and proposals grounded in these foundations, and surface gaps and new questions.
Business questions, data definitions, calculations and business rules are preserved with their versions. Results, expert assessments and decisions are linked to these foundations. Each contribution retains its authorship and the conditions in which it was produced.
Actual outcomes are compared with expectations using data. People and AI solutions revisit the findings. Calculations, business rules, assumptions or practices are revised where needed.
THE QUESTION GUIDING MY WORK
How can the solutions and decisions we develop together
improve the organization’s experience?
02 / FOUR PERSPECTIVES
An evaluation framework, a theory of experiential learning, an approach to organizational learning and a model of knowledge creation.
EXPERIENTIAL LEARNING
Kolb’s approach explores learning through the relationship between concrete experience, reflective observation, abstract conceptualization and active experimentation.
In my workReviewing the outcome of a practice, developing a method from it and testing that method in a new situation.
Explore the source ↗SINGLE- AND DOUBLE-LOOP LEARNING
Argyris and Schön’s approach to organizational learning distinguishes correcting errors within existing rules from questioning the assumptions that govern those rules.
In my workWhen a decision does not produce the expected outcome, revisiting the objective, business rule and underlying assumption as well as the action.
Reference details ↓ORGANIZATIONAL KNOWLEDGE CREATION
Nonaka’s SECI framework explores the conversion between tacit and explicit knowledge through socialization, externalization, combination and internalization.
In my workHelping experts articulate experience as methods, and supporting the further development of those methods through practice.
Original paper ↗FOUR LEVELS OF EVALUATION
The Kirkpatrick framework considers evaluation at the levels of reaction, learning, behavior and results. The transfer of learning to workplace behavior and intended outcomes is particularly relevant to my work.
In my workDefining expected behavior, the baseline and relevant business indicators before training. Avoiding the assumption that a change in results is, on its own, evidence of training impact.
Official model overview ↗03 / FROM THOUGHT TO PRACTICE
Data-backed methods and shared decision experience form the common foundation of my work at Telkraft and Promptever.
CONSULTING PRACTICE
We work with organizational experts to define the data, calculations and business rules needed for critical business questions. We carry out this work without connecting to the organization’s data systems and deliver reusable recipes, methods and code for use within its own systems. AI use is optional.
Explore Telkraft ↗
PrompteverPRODUCT DEVELOPMENT
Promptever, which I founded, preserves defined problem-solving methods and decision experience in a shared memory. Without connecting data or using AI, you can structure methods and record and revisit findings, forecasts and decisions. When data systems are connected, deterministic calculation results provide the memory’s evidence base. People and any AI tools you choose to use can share and develop the same methods and past context.
Promptever is part of the Türkiye AI Initiative (TRAI) ecosystem.
Explore Promptever ↗04 / ESSAYS & NOTES
Notes on organizational learning. I will develop this space with longer essays and explorations over time.
A decision record alone does not explain the conditions in which the decision was made. Considering the method, available knowledge, assumptions and expected outcome together makes the experience open to review.
For me, the learning question begins here: if the actual outcome differs from our expectations, do we change only the action, or also question the assumption behind the decision?
Practice perspective · A connection to Argyris–Schön
How an expert reaches a good outcome is not always written down. The questions they ask, the data they distrust and the conditions that lead them to change a method are also part of their expertise.
Structuring this knowledge can support collaboration. Yet writing a recipe does not mean that all tacit knowledge has been transferred. Joint practice, feedback and new experiences sustain knowledge creation.
Practice perspective · A connection to Nonaka · SECI
A performance indicator may improve after training. Processes, equipment, workloads or team composition may also have changed in the same period. Temporal association alone does not establish causation.
This is why I value framing the evaluation question before training. The baseline, expected behavior, practical conditions and, where possible, an appropriate comparison should be designed together.
Practice perspective · A focus on Kirkpatrick L3–L4
CAN GÖKÇAY05 / ABOUT ME
I have worked in technology, product development and EdTech for more than 25 years. As a CTO, technical product manager and project manager, I have led the development of more than 50 national and international products and projects across education, healthcare, the public sector and enterprise technology.
Today, my work focuses on how people and AI solutions can contribute to organizational learning together through organizational data, deterministic calculations and shared memory. I explore how organizations can turn data-backed results into problem-solving and decision experience. I am co-founder and CTO at Telkraft, and founder and product architect at Promptever.
This site is my personal workspace for bringing together readings, reflections and practical explorations on organizational learning.
Get in touch on LinkedIn ↗Model descriptions are summaries. The notes and connections to practice reflect my own perspective.