Lead Software Engineer and Product Owner building industrial IoT and data products from edge devices to analytical models and user-facing applications.
At OTOH, I connect business strategy, hands-on engineering, and product management. I translate business goals and customer needs into clear requirements, prioritize deliverables by impact, technical feasibility, and expected ROI, and track the roadmap and KPIs while designing and operating software across embedded systems, backend, data, and frontend.
My path started in acoustic R&D in 2020 and evolved into production IoT work and open-source data engineering across embedded Python, event-driven backends, sensor data, analytical modeling, and product development.
- Data and analytics: Python, SQL, dbt, Airflow, PostgreSQL, TimescaleDB, dimensional modeling, data quality, lineage, and time-series analytics.
- Industrial IoT: Raspberry Pi, Python, digital signal processing, MQTT 5/TLS, AWS IoT Core, Device Shadow, and edge buffering.
- Software engineering: event-driven backends, FastAPI, Node.js, React, TypeScript, Next.js, industrial dashboards, and idempotent pipelines.
- Product ownership: business and customer requirements, expected ROI, discovery, roadmap, KPIs, technical architecture, and hands-on delivery across hardware, software, data, and user experience.
A reproducible analytics-engineering pipeline that turns NASA IMS bearing-vibration snapshots into validated, auditable analytical data and a Tableau dashboard for retrospective condition investigation.
- Typed Python, NumPy, and SciPy feature engineering with explicit quality flags
- Idempotent PostgreSQL persistence with source lineage, checksums, and reprocessing history
- Tested dbt dimensional models with baseline comparisons, rolling metrics, and a deterministic BI contract
- Airflow orchestration, Docker-based infrastructure, automated tests, and GitHub Actions CI
- Verified processing of 984 snapshots into 3,936 traceable feature rows
An edge-to-cloud platform connecting Modbus TCP equipment, Node-RED, MQTT QoS 1, a Python ingestion worker, TimescaleDB, FastAPI, and a Next.js dashboard.
- Temporal sensor replacement with non-overlapping assignment history
- Idempotent telemetry ingestion and explicit data-quality handling
- Alert hysteresis and resolution-routed time-series queries
- Transactional MQTT outbox and JWT/OIDC authentication
- Automated backend and frontend tests, Docker, and CI/CD
Live demo · Documentation · API · Video
A Python sound-level meter developed from my Acoustical Engineering work with Raspberry Pi and digital signal processing.
Python · SQL · PostgreSQL · dbt · Airflow · Tableau · TimescaleDB · FastAPI · MQTT · Docker · AWS · TypeScript · React · Next.js