Capability taxonomy and evidence map.

Canonical Gutruf role evidence for wireless battery-free photometry and optical bioelectronics.

• Developed fully implantable wireless and battery-free platforms for neural recording, neurostimulation, musculoskeletal monitoring, and functional electrical stimulation.
• Built reusable validation infrastructure across 20+ test fixtures, 6 tuned antenna designs, and 3 simulation frameworks.
• Contributed to peer-reviewed work in PNAS, Microsystems & Nanoengineering, and Nature Communications, with 14 peer-reviewed journal articles.

Canonical project evidence for optical source/detector hardware, photometry, and miniaturized implant integration.

• Optical source/detector hardware, miniaturized electronics, wireless power, flexible probe mechanics, communication, implant packaging, and validation workflows.
• The platform connected optical readout, flexible probe placement, wireless power, and chronic packaging into a small subdermal system.
• Miniaturized optical implants need mechanical, optical, wireless, and surgical constraints resolved together.

Canonical Rhaeos role evidence for wearable sensing, reliability, charging, adhesives, and daily use.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for clinical and home-use sensing workflows.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical project evidence for home monitoring, charging, placement, and adhesive iteration.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical project evidence for adapting the wearable sensing stack into a soft wound-monitoring patch.

• Sensor selection, adhesive fabrication, product design, testing, thermal-transfer simulation, diffusivity back-calculation, preclinical data collection, and multimodal analysis workflows.
• Thermal response, humidity dynamics, and peripheral temperature features created a quantitative path from raw patch signals to wound-healing progression.
• The best sensing platforms are built with the study workflow and feature model from the beginning, not after hardware is finished.

Canonical role evidence for bench, field, clinical, home, and data-review validation work.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for algorithm validation and FDA-support workflows.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical project evidence for home-use study support and workflow refinement.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical project evidence for preclinical wound-monitoring analysis.

• Sensor selection, adhesive fabrication, product design, testing, thermal-transfer simulation, diffusivity back-calculation, preclinical data collection, and multimodal analysis workflows.
• Thermal response, humidity dynamics, and peripheral temperature features created a quantitative path from raw patch signals to wound-healing progression.
• The best sensing platforms are built with the study workflow and feature model from the beginning, not after hardware is finished.

Canonical role evidence for integrating electronics, packaging, adhesives, charging, and tests.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for tying thermal sensing physics to model validation.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical project evidence for sensor selection and prototype fabrication.

• Sensor selection, adhesive fabrication, product design, testing, thermal-transfer simulation, diffusivity back-calculation, preclinical data collection, and multimodal analysis workflows.
• Thermal response, humidity dynamics, and peripheral temperature features created a quantitative path from raw patch signals to wound-healing progression.
• The best sensing platforms are built with the study workflow and feature model from the beginning, not after hardware is finished.

Canonical role evidence for clinical and home monitoring workflows.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for the clinical sensing workflow.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical project evidence for daily-life and sleep monitoring.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical role evidence for wavelet and DSP methods in neural-signal interpretation.

• Built and optimized front-end hardware, phantoms, and signal-processing workflows for 4D acoustoelectric current-density imaging.
• Presented the non-invasive neural-recording work at BMES 2017 and IEEE IUS 2018.

Canonical role evidence for calibration, labeling, prediction, and time-series data.

• Extracted ECG, HRV, posture, and activity features from 800+ hours of chest-worn wearable data across 31 subjects.
• Built wearable physiological-data workflows spanning filtering, feature generation, calibration, labeling, and prediction support.

Canonical role evidence for sensor review, algorithm evaluation, and device-performance analysis.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for feature engineering and algorithm validation.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical role evidence for calibration, labeling, and prediction workflows.

• Extracted ECG, HRV, posture, and activity features from 800+ hours of chest-worn wearable data across 31 subjects.
• Built wearable physiological-data workflows spanning filtering, feature generation, calibration, labeling, and prediction support.

Canonical project evidence for data organization and thermal visualization.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical role evidence for device data collection, review, and algorithm evaluation.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for product-relevant model-training and feature workflows.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical project evidence for home-use data organization.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical role evidence for structured physiological-data handling.

• Extracted ECG, HRV, posture, and activity features from 800+ hours of chest-worn wearable data across 31 subjects.
• Built wearable physiological-data workflows spanning filtering, feature generation, calibration, labeling, and prediction support.

Canonical role evidence for amplifier optimization and EMI control.

• Built and optimized front-end hardware, phantoms, and signal-processing workflows for 4D acoustoelectric current-density imaging.
• Presented the non-invasive neural-recording work at BMES 2017 and IEEE IUS 2018.

Canonical project evidence for implantable-system instrumentation constraints.

• Wireless photometry, neurostimulation, osseosurface sensing, and FES systems using miniaturized circuits, flexible interconnects, soft packaging, and biocompatible encapsulation.
• The reusable engineering base made it possible to translate wireless power, communication, packaging, and validation methods across multiple implantable publications.
• Implantable engineering succeeds when power, mechanics, packaging, test fixtures, and preclinical workflow are designed as one system.

Canonical role evidence for implantable wireless and battery-free bioelectronics.

• Developed fully implantable wireless and battery-free platforms for neural recording, neurostimulation, musculoskeletal monitoring, and functional electrical stimulation.
• Built reusable validation infrastructure across 20+ test fixtures, 6 tuned antenna designs, and 3 simulation frameworks.
• Contributed to peer-reviewed work in PNAS, Microsystems & Nanoengineering, and Nature Communications, with 14 peer-reviewed journal articles.

Canonical project evidence for fully implantable wireless systems.

• Wireless photometry, neurostimulation, osseosurface sensing, and FES systems using miniaturized circuits, flexible interconnects, soft packaging, and biocompatible encapsulation.
• The reusable engineering base made it possible to translate wireless power, communication, packaging, and validation methods across multiple implantable publications.
• Implantable engineering succeeds when power, mechanics, packaging, test fixtures, and preclinical workflow are designed as one system.

Canonical role evidence for wireless power in implantable systems.

• Developed fully implantable wireless and battery-free platforms for neural recording, neurostimulation, musculoskeletal monitoring, and functional electrical stimulation.
• Built reusable validation infrastructure across 20+ test fixtures, 6 tuned antenna designs, and 3 simulation frameworks.
• Contributed to peer-reviewed work in PNAS, Microsystems & Nanoengineering, and Nature Communications, with 14 peer-reviewed journal articles.

Canonical role evidence for wireless charging in wearable medical devices.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for charging iteration in the home wearable.

• Wearable builds, adhesive/sensor layout iterations, modular electronics/sensor/battery architecture, Qi charging, motion sensing, onboard memory, data encryption, remote collection, and thermal visualization scripts.
• Reliability improved because patient, caregiver, clinician, software, and hardware feedback directly informed form factor, placement, charging, adhesive, and data-quality decisions.
• Home monitoring is an integrated system problem: sensor performance, usability, reliability, and remote workflows have to improve together.

Canonical role evidence for flexible, soft, and biocompatible platforms.

• Developed fully implantable wireless and battery-free platforms for neural recording, neurostimulation, musculoskeletal monitoring, and functional electrical stimulation.
• Built reusable validation infrastructure across 20+ test fixtures, 6 tuned antenna designs, and 3 simulation frameworks.
• Contributed to peer-reviewed work in PNAS, Microsystems & Nanoengineering, and Nature Communications, with 14 peer-reviewed journal articles.

Canonical project evidence for flexible circuits, soft materials, and encapsulation.

• Wireless photometry, neurostimulation, osseosurface sensing, and FES systems using miniaturized circuits, flexible interconnects, soft packaging, and biocompatible encapsulation.
• The reusable engineering base made it possible to translate wireless power, communication, packaging, and validation methods across multiple implantable publications.
• Implantable engineering succeeds when power, mechanics, packaging, test fixtures, and preclinical workflow are designed as one system.

Canonical project evidence for soft-device work in a wearable patch.

• Sensor selection, adhesive fabrication, product design, testing, thermal-transfer simulation, diffusivity back-calculation, preclinical data collection, and multimodal analysis workflows.
• Thermal response, humidity dynamics, and peripheral temperature features created a quantitative path from raw patch signals to wound-healing progression.
• The best sensing platforms are built with the study workflow and feature model from the beginning, not after hardware is finished.

Canonical role evidence for preclinical validation in freely moving animal models.

• Developed fully implantable wireless and battery-free platforms for neural recording, neurostimulation, musculoskeletal monitoring, and functional electrical stimulation.
• Built reusable validation infrastructure across 20+ test fixtures, 6 tuned antenna designs, and 3 simulation frameworks.
• Contributed to peer-reviewed work in PNAS, Microsystems & Nanoengineering, and Nature Communications, with 14 peer-reviewed journal articles.

Canonical project evidence for preclinical implantable systems.

• Wireless photometry, neurostimulation, osseosurface sensing, and FES systems using miniaturized circuits, flexible interconnects, soft packaging, and biocompatible encapsulation.
• The reusable engineering base made it possible to translate wireless power, communication, packaging, and validation methods across multiple implantable publications.
• Implantable engineering succeeds when power, mechanics, packaging, test fixtures, and preclinical workflow are designed as one system.

Canonical project evidence for preclinical wound-monitoring analysis.

• Sensor selection, adhesive fabrication, product design, testing, thermal-transfer simulation, diffusivity back-calculation, preclinical data collection, and multimodal analysis workflows.
• Thermal response, humidity dynamics, and peripheral temperature features created a quantitative path from raw patch signals to wound-healing progression.
• The best sensing platforms are built with the study workflow and feature model from the beginning, not after hardware is finished.

Canonical role evidence for requirements, test procedures, travelers, records, and design notes.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical project evidence for FDA-support documentation and validation tooling.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical independent-work evidence for private AI, retrieval, and automation workflows.

• Built a private evidence and retrieval workflow spanning structured facts, source links, OCR, embeddings, and search.
• Connected local infrastructure, AI tooling, and automation into a maintainable engineering knowledge system.

Canonical project evidence for local AI-agent workflow systems.

• A local agent-oriented setup with project notes, memory integration, infrastructure hooks, and a workflow for moving from idea to implementation.
• The strongest part was using the system as an engineering workspace rather than a generic chatbot wrapper.
• AI tools are most useful when they are constrained around a real workflow, not when they become another cluttered interface.

Canonical ACE project evidence for DSP/ML validation, explainability, and deployment-check workflows.

• Automated ingestion, provenance, labeling, QC, dataset versioning, feature generation, validation, bias checks, explainability, and deployment-check workflows.
• The strongest result was connecting sensor physics, clinical usability, small-dataset controls, and deterministic deployment checks into one regulated workflow.
• Physiological ML is strongest when feature design, validation, and deployment checks stay grounded in sensing physics and clinical workflow constraints.

Canonical Rhaeos role evidence for ML validation workflows used in wearable medical-device development.

• Improved FlowSense Home field reliability from about 70% in Gen 1 to about 96% by the final Gen 3 clinical study.
• Led 20–30 clinical-study device builds and reduced build cycles from about 6 weeks to about 3 weeks.
• Advanced FlowSense Home across 3 design generations, 100+ manufactured units, 200+ participants, and 2,800+ home-device wear hours.

Canonical independent-work evidence for local-first infrastructure, file services, and backup/recovery.

• Built a private evidence and retrieval workflow spanning structured facts, source links, OCR, embeddings, and search.
• Connected local infrastructure, AI tooling, and automation into a maintainable engineering knowledge system.

Canonical project evidence for storage, media, AI services, networking, and remote access.

• A Proxmox-based system with storage, media services, Nextcloud, Immich, DNS, local AI services, and remote access.
• The system became useful because it solved real needs: storage, media, backups, AI experiments, and remote access.
• Infrastructure projects need the same discipline as product work: simple entry points, clear ownership, backups, and documentation.

Canonical project evidence for local infrastructure and integration hooks.

• A local agent-oriented setup with project notes, memory integration, infrastructure hooks, and a workflow for moving from idea to implementation.
• The strongest part was using the system as an engineering workspace rather than a generic chatbot wrapper.
• AI tools are most useful when they are constrained around a real workflow, not when they become another cluttered interface.

Canonical project evidence for storage, networking, and remote access.

• A Proxmox-based system with storage, media services, Nextcloud, Immich, DNS, local AI services, and remote access.
• The system became useful because it solved real needs: storage, media, backups, AI experiments, and remote access.
• Infrastructure projects need the same discipline as product work: simple entry points, clear ownership, backups, and documentation.