Lead AI Engineer at BeWell Clinical Studies. I research and build regulatory-compliant AI systems for clinical research (HIPAA, NIST 800-171, 21 CFR Part 11), and open-source hardware on the side.
Twelve years in healthcare IT before the AI lead role: sole IT for a 45-person organization, infrastructure and compliance from the ground up, and a long arc of teaching myself into LLM deployment, agentic systems, and embedded ML. I think in systems and ship them.
A Raspberry Pi Zero 2 W you wear that captures audio, runs local STT, and stores notes, sessions, and activity in a queryable knowledge graph. An ESP32-S3 USB dongle handles BLE fallback when WiFi isn't available, an MCP server lets Claude and other agents read and write the memory directly, and it now runs in the cloud as a private, single-owner service.
- cortex-core is the Pi firmware and SQLite backend
- Cortex-Cloud is the one-command Azure deployment: a private web memory Hub plus MCP for any AI, locked to a single account
- cortex-link is the ESP32-S3 BLE bridge with OTA via ugit
- cortex-mcp is the MCP server and CLI for AI agents
- cortex-desktop is the system-tray companion app
- cortex-plugin is the Claude Code plugin
- cortex-voice-training fine-tunes Whisper on Cortex recordings
A building accessible prosthetic sensing. Pressure myography (PMG) and tissue deformation myography (TDMG) for finger tracking, no EMG required, using custom flex PCBs and real-time ML.
- Open Muscle Hub is the central docs and build guides
- FlexGrid is a 60-sensor wearable pressure array
- LASK5 is a handheld gesture training device
- OpenMuscle Software is the firmware, data tools, and ML hooks
- Models on Hugging Face · writeup at openmuscle.org · featured on Hackaday
The original turfptax/openmuscle proof-of-concept repo is archived. Active development continues across the Open Muscle GitHub organization linked above.
A small library for over-the-air firmware updates from a GitHub repo. Widely used in the ESP32/MicroPython community.
- bugout-monitor is a disaster-prep dashboard with OSINT threat monitoring and LLM integration
- offline-mesh-toolkit is a download-once, flash-anywhere toolkit for Meshtastic and MeshCore
- anti-gaslight-firmware is an ESP32-S3 environmental sensor (BME280 + TCS3200 + PIR to MQTT) with OTA via ugit
- ESP32Watch is MicroPython watch firmware for the Waveshare ESP32-S3 AMOLED-2.06
- openclaw-udp-messenger is a local-network agent-to-agent messaging plugin
Auto-updated 2026-08-03 · top 5 most recently pushed repos
Cortex-Cloud·Python- Your own AI memory, in the cloud you own. Deploy a private, single-owner AI memory service (web Hub + MCP for…cortex-core·Python- Cortex Core: Pi-side memory + overseer plugin. Local-first wearable AI memory: notes, sessions, projects, tim…cortex-desktop·TypeScript- Cortex Hub desktop app: system tray + browser UI for your AI companion Piturfptax.github.io·HTML- (no description)cortex-plugin·Python- Claude Code plugin for the Cortex wearable AI memory system
See also: recent metrics digest.
AI: LLM deployment, RAG, agentic frameworks, MCP servers, regulatory-compliant pipelines for clinical and healthcare contexts Embedded: ESP32-S3 / ESP32-C3, Raspberry Pi, MicroPython (asyncio), custom flex PCBs via JLCPCB, sensor fusion, ESP-NOW ML: Real-time inference, signal processing, hand-gesture recognition, fine-tuning Whisper for domain-specific STT Tooling: Python, JavaScript/Node, KiCad, Fusion 360, Blender
I work as an applied AI researcher and integrator. I deploy, orchestrate, and harden real systems in production, and I build novel ones (wearable AI memory, open-source biosensing) to find out what these tools can actually do.
Sci-fi writer, occasional YouTube creator at TURFPTAx, and a slow-burn fan of any system with a sharp learning curve. I read a lot.
I'm open to work on AI in regulated industries, biosensing, embedded ML, and assistive technology. Beginner-friendly issues are marked across the Open Muscle repos, so pull up a chair.
More at openmuscle.org.