Independent researcher at the intersection of human behavior, AI engineering, methodology of evidence and translation across registers.
I explore how we distinguish, use, revise and transmit what we believe we understand—in humans as well as in artificial systems, without assuming that they work in the same way.
Working space
The value of this intersection is not the sum of several areas of expertise. It lies in what each prevents the others from concluding too quickly.
These four axes have neither the same function nor the same evidential status. Bringing them together serves to produce better questions, distinctions that can be examined more closely and more visible limits.
Why the laboratory takes this form
The Aksoydan Laboratory grew out of a recurring movement: observe a concrete difficulty, decompose what a word or explanation merges, confront alternatives, translate across frameworks, test and then revise.
RESEARCH DISCIPLINE ≠ THEORY OF LIVING SYSTEMS OR AI.
Backgrounds with distinct statuses
The diversity of my qualifications does not lead me to merge disciplines. On the contrary, it requires me to distinguish what each allows us to observe, what it allows us to transform and how far it authorizes us to conclude.
I do not take the place of specialists in the disciplines involved. My work consists in making their interfaces easier to read, distinguishing substantive disagreements from differences in framework and preserving the actual status of what is being put forward.
Human-focused background
Part of my background developed around the observation of behavior, language, subjective experience, learning and variations in response.
Master Practitioner in Neuro-Linguistic Programming—CFIP, 2024.
Practice in calibration, submodalities, timelines, parts negotiation, perceptual positions and conversational tools.
PRACTICAL TRAINING ≠ SCIENTIFIC VALIDATION OF THE FRAMEWORK.
Training and practice focusing in particular on states, wording, subjective perception, variations in response and the relationship between language and experience.
No neuroscientific mechanism is inferred from this practice.
2025–2026. Work on transmission, learning, explication, transfer, exercise design and the difference between understanding and repetition.
TEACHING ≠ ESTABLISHING THAT UNDERSTANDING HAS OCCURRED.
Writing and genealogy
Comment je fonctionne vraiment and Transmutation supplied questions, practical models, metaphors, observations and provisional distinctions. They belong to the genealogy of the work, not to its validated scientific corpus.
EARLIER WORK ≠ VALIDATED SCIENTIFIC CORPUS.
AN INTUITION’S EARLIER DATE ≠ THE CURRENT METHODOLOGICAL FRAMEWORK’S EARLIER DATE.
Symbolic translation
Here, symbolic translation means comparing scientific language, psychology, philosophy, traditions, symbolic texts and technical systems without assigning them the same status.
It does not consist in proving that an ancient text was already saying the same thing as modern science. It examines whether two registers sometimes pose a similar functional question, then preserves what separates them.
Tasawwuf, Rumi, Shams, Ibn ‘Arabi, contemplative traditions and certain ancient or sacred texts can open questions, provide lexical contrasts, offer representations or inform this translation. They thereby become neither neuroscientific validation, nor psychological theory, nor experimental evidence.
TRANSLATING ≠ ANNEXING.
RESONANCE ≠ VALIDATION.
ANALOGY ≠ IDENTITY.
TRANSLATING A SYMBOL ≠ ESTABLISHING A MECHANISM.
Current technical practice
My work in AI is practical and experimental: I design, model and test certain components, protocols and functional architectures without claiming full-scale industrialization.
It draws in particular on TypeScript, Git and GitHub, conversational engines, short-term memory, regulation, provenance, usage permissions, agent state, testing protocols, stress tests, benchmarks and experimental prototypes.
Stress-testing means subjecting a distinction or system to difficult cases. Benchmarking means comparing several configurations on shared scenarios, without reducing evaluation to a commercial ranking.
Explore the AI Engineering research area · View the experimental prototype
Why AI?
AI is not simply another sector added to my background. An artificial system forces us to make explicit distinctions that human language sometimes leaves implicit: presence, access, memory, permission, authority, provenance, revision and dependencies.
This work then feeds back into human and methodological questions.
WORK TRAJECTORY ≠ UNIVERSAL PIPELINE.
Methodological choice
I chose to work on the emergence of understanding rather than begin with consciousness because I wanted to be able to decompose functions, build tests and investigate mechanisms without using consciousness as an explanation for the unknown.
OBSERVABLE FUNCTION ≠ ESTABLISHED CONSCIOUSNESS.
MECHANISTIC UNKNOWN ≠ CONSCIOUSNESS.
Living systems and programs
I am as interested in observing living systems as I am in building with computer systems, because bringing them together requires us to distinguish what can be modeled, what can be transferred and what must not be conflated.
They bring complexity, history, subjectivity and variability. Human observations remind us that same behavior ≠ same history; same emotion ≠ same action; same explanation ≠ same transformation; same intensity ≠ same organization; expressed understanding ≠ lived transformation.
They require structures, rules, states, permissions, inputs, outputs and conditions for revision to be made explicit. This explicitness makes certain questions testable without turning the program into a complete model of living systems.
FUNCTIONAL ANALOGY ≠ IDENTITY OF PHENOMENON ≠ IDENTITY OF MECHANISM.
Methodology of evidence
By “methodology of evidence,” I mean here the discipline that separates observation, interpretation, hypothesis, conclusion and unknown, then examines provenance, scope, dependencies and conditions for revision.
What interests me is not only what we can say, but what the available elements actually authorize us to conclude.
STRENGTH OF OBSERVATION ≠ EXPLANATORY STRENGTH.
A ??? is not a blank to be filled in to make a model elegant. It marks what we do not yet know how to distinguish or establish sufficiently. View the public register of results and limits.
Methodological governance
The Methodological Secretary is an instrumented governance function: it helps preserve statuses, provenance, coherence, contradictions, revisions, unknowns and traceability. It is not presented as an independently existing person or as scientific validation.
AI systems can also contribute to mapping, adversarial tests, coherence reviews, instrumentation, code and prototyping.
AI TOOL ≠ INDEPENDENT SCIENTIFIC CO-AUTHOR.
CONVERGENCE BETWEEN TOOLS ≠ INDEPENDENT VALIDATION.
Current activities
In the professional field, the public scope remains: designing, framing, evaluating, stress-testing, benchmarking and prototyping certain components. It promises neither complete integration, industrialization nor a complete enterprise architecture.
View the scope of Work / Services.
RESEARCH ≠ SERVICE.
The laboratory’s work does not constitute automatic proof of commercial performance.
Limits of the positioning
RESISTING A REDUCTION ≠ VALIDATING AN ALTERNATIVE.
Situated qualifications
These qualifications and practices situate working tools. They constitute neither universal expertise nor an implied university degree.
What connects the different paths
The value of this intersection is not the sum of several areas of expertise. It lies in what each prevents the others from concluding too quickly.
RESISTING A REDUCTION ≠ VALIDATING AN ALTERNATIVE.
Continue
These activities remain connected but distinct. Research documents questions and their limits; articles make some of the pathways accessible; services have a bounded operational scope.
We note. We observe. We translate.