Essays
Arguments and architectural positions that need enough room to expose their pressure points.
Research surface / build notes / pressure tests
Build before explanation.
I work on increasingly capable machine systems and the harder question underneath them: how to make capability useful without confusing capability with authority.
This is the public research surface around that work. Candidates can be strange. Claims still have to survive contact with artifacts, falsifiers, and adverse evidence.
01 / Thesis
The machine can get stronger without becoming sovereign.
As AI systems become more persistent, agentic, and operationally capable, the central architectural problem shifts. The question is no longer only what the model can do. It is what the surrounding system permits it to do, on whose authority, with what resources, and with what evidence left behind.
I am interested in the connective tissue between competence and consequence: admissibility before execution, operator authority, resource governance, provenance, reviewability, and safe actuator boundaries.
Capability is a property of the worker. Authority is a property of the system around it.
02 / Method
Discovery gets range. Execution gets friction.
Prematurely forcing every insight into a practical next step is a good way to erase discoveries that do not yet fit existing categories. Exploration needs room for mutation, unusual comparisons, and candidate hypotheses.
That freedom stops at the execution boundary. Code promotion, deployment, communication, spending, confidential access, and other external effects require stronger evidence and explicit authority.
01
Generate candidates aggressively.
Do not confuse unfamiliarity with uselessness.
Do not confuse unfamiliarity with uselessness.
02
Preserve state distinctions.
Specified, implemented, locally validated, externally proven, and speculative are different states.
Specified, implemented, locally validated, externally proven, and speculative are different states.
03
Keep adverse evidence.
A blocked run or falsified claim is part of the system's knowledge, not debris to hide.
A blocked run or falsified claim is part of the system's knowledge, not debris to hide.
04
Do not let the worker self-authorize.
Model confidence is not an admissibility decision.
Model confidence is not an admissibility decision.
03 / Surfaces
Read by function, not by chronology.
Notes
Short propositions, live candidates, and observations that may remain unresolved.
Evidence
Claim-state discipline, public sources, explicit non-claims, and inspectable surfaces.
About
The founder position, the relationship to ANANS, and the publication boundary.
04 / Field
Current lines of inquiry.
- Persistent machine systems.
Software that perceives changing environments, reasons across longer horizons, coordinates work, and remains inspectable. - Agent-to-agent infrastructure.
What changes when machine systems become continuous participants in work rather than tools opened for isolated tasks. - Authority architecture.
How execution rights, operator authorization, budgets, provenance, and reviewability remain external to model confidence. - Endogenous discovery.
How systems can generate, stress, compare, falsify, and preserve candidates without collapsing discovery into truth. - Governed build systems.
How coding agents and autonomous work loops can be useful without becoming their own admission authority.
05 / ANANS
The build programme is separate from the publication.
ANANS is the active research-and-build programme through which I test many of these questions in software. It is not reducible to an AI governance tool, compliance product, agent framework, DevSecOps system, or consulting method.
Undomesticated AI is not an ANANS mirror. It is the founder/research surface. Public ANANS claims should point toward inspectable evidence where possible; private runtime, trade-secret material, and confidential research remain outside this site.
06 / Non-claim
Work in public without pretending the work is finished.
Nothing on this site should be read as evidence that ANANS is finished, externally validated, production-ready, or generally autonomous. A working implementation remains a working implementation. A synthetic demonstration remains synthetic. An external claim remains external until source-bound.
The publication can move fast at the level of ideas while staying conservative about what has actually been proven.