Build. Test. Question.
A selective view of experimental software and technical inquiry. We report working mechanisms, measured findings, and limitations separately. Source implementations remain private unless deliberately released.
Resident Model Lab
Local OLMoE inference with router-faithful expert residency. Compare the v0.2 memory, latency and fidelity findings with their experimental limits.
Read Resident Model Lab methods and limitations →Research Stack
Public research software v0.1.0: TRACE · LOOP · LENS · FRAME connects provenance, evaluation, visualization and reviewed corrections. Includes NRS-Bench, three technical reports and 54 developer/CI-reported tests on controlled synthetic evaluations, not a real-model or human study.
Read Research Stack methods and limitations →VEYDRIS
Recorded-video detection, temporal tracking, replay and exports in a local prototype. The synthetic map is not verified geolocation.
Read VEYDRIS methods and limitations →Human–AI interaction & explainable AI
Evidence provenance, conversational memory and inspectable uncertainty, with founder-led Epistemic Trust Lab synthetic-policy evaluation kept distinct.
Read Human–AI interaction & explainable AI methods and limitations →- /01Perception
Sensing & classification utilities
Small, non-sensitive examples of signal handling and object classification pipelines.
- /02Inference
Evaluation & orchestration patterns
Reference patterns for evaluating models and composing intelligent workflows.
- /03Oversight
Human-in-the-loop interfaces
Interface studies for reviewing machine recommendations before action.
- /04Explainability
Inspectable AI Interfaces
Interface patterns for transparent model explanations, uncertainty, evidence provenance, and human correction or override.
Company research is described here without disclosing unpublished source code or private experimental materials.