Myant is developing a smart clothing platform, which connects the human operating system to the cloud. They call this textile computing.
At Myant, I led all software engineering efforts, ranging from native iOS/Android applications, web based GraphQL API’s, to streaming biometric ingestion engines. I led a team of software engineers distributed between Ottawa and Toronto, and worked closely with UX/UI designers, machine learning specialists, and hardware/firmware engineers.
I also championed an in house biometric processing engine:
•Functional reactive SDK, designed for real-time stream-processing
•Single python process, designed to be run behind a websocket, with minimal latency
•Snapshot and restore capabilities for stateful (i.e. windowed) algorithms
The SDK allowed our machine learning specialists to write machine learning and signal processing algorithms in pure python, while abstracting away the broader system. Data would be collected via sensors embedded in an array of clothing, transmitted via Bluetooth to a mobile app, where it was aggregated and streamed to a cloud environment via a websocket. Algorithm outputs (heart rate, breathing rate, stress, etc.) would be streamed back to the application in real-time. High density data (ex ECG at 1000 Hz) would be serialized, and saved to a custom federated database (Redis, PostgreSQL, S3)