Skills

Stack behind Renata platforms and open-source analytics tooling. Each line says where it shows up — not a generic chip list. Sample projects link to public GitHub where available.

languages

• Python
  Pipelines, agents, and shop-floor tooling
  Sample Projects (3)

  • databao-agent
  • quak
  • analytics-agent-duckdb-workshop
• TypeScript
  MCP servers and production web apps
  Sample Projects (1)

  • mcp-server-motherduck
• SQL
  Warehouse truth layer for forensics and agents
  Sample Projects (2)

  • mcp-server-motherduck
  • databao-agent
• Rust
  Performance-sensitive tooling when it earns its keep
• Go
  Services and CLI utilities where static binaries help

industrial

• MQTT
  Renata production messaging path — PLC/SCADA to analytics
• EMQX
  Broker topology and access for battery manufacturing IIoT
• OPC UA
  Machine connectivity under OT isolation constraints
• SCADA
  Shop-floor supervision systems that cannot go down lightly
• PLC
  Line control reality — where uptime and change risk meet

data

• DuckDB
  ADDA warehouse and open-source analytics agent stacks
  Sample Projects (3)

  • mcp-server-motherduck
  • quak
  • analytics-agent-duckdb-workshop
• MotherDuck
  Cloud DuckDB path with agent-safe access patterns
  Sample Projects (1)

  • mcp-server-motherduck
• PostgreSQL
  Transactional store for apps that need ACID, not just analytics
• MongoDB
  Document stores when schema flexibility beats rigid tables
• Apache Spark
  Batch / large-scale transforms when DuckDB is not the right tool

frontend

• React
  Investigation and ops UIs used by quality and production
• Next.js
  App shell for ADDA-class manufacturing frontends
• Svelte
  This CV site and lean UI surfaces

ai

• MCP
  Tooling with explicit DB boundaries — useful before AI meets plant data
  Sample Projects (2)

  • mcp-server-motherduck
  • analytics-agent-duckdb-workshop
• LLMs
  Agents grounded in SQL truth, not slideware demos
  Sample Projects (1)

  • databao-agent

cloud

• Netlify
  Public properties — deploy, DNS, TLS, and support path
• AWS
  Cloud services when the workload leaves on-prem
• Azure
  Enterprise / Swatch-adjacent cloud landscapes
• GCP
  Select workloads and data services as needed

Also

databases: DuckDB, DuckLake, MotherDuck, PostgreSQL, MongoDB, SQLite

cloud: Netlify, AWS, Azure, GCP

How I work

• Shop-floor fluency — Twenty-plus years in automation and manufacturing — I know what breaks at 2 a.m. and who owns the risk.
• Hands-on leadership — Head of OT, IT & ISM who still writes schemas, ingest, SQL, UI, and deploy paths.
• OT / IT bridge — Connect production engineering, quality, and security around systems that actually go live.
• Schema-first platforms — Design storage and query layers engineers can trust for forensics and production analytics.
• Secure IIoT pragmatism — MQTT/EMQX and access patterns aligned with an IEC 62443 mindset under real production constraints.
• Regulated AI tooling — MCP and analytics agents with explicit database boundaries — useful without fantasy demos.
• End-to-end ownership — From line sensors to public properties — UX, content, DNS, TLS, and production support.
• Open source practice — Publish tools you rely on (mcp-server-motherduck, databao-agent, workshops).