Industrial edge systems that survive real operations

Device fleets, edge data pipelines, and AI-powered operator tooling

I run edge data-logging devices in production across industrial sites — Modbus on the machine side, containerized collectors, one-click OTA updates — and I build the AI workflow tooling that makes systems like that observable and safe to operate. Boring, reliable choices over interesting ones, because these systems have to work for years.

Edge Data Logging

Fleet OTA Updates

AI Operations Tooling

Operating Contexts

Industrial ReliabilityEdge Device FleetsOperator ConsolesKnowledge Systems

View Projects Get in Touch

Background, CV, and career context live on the About page.

The Agent Stack

Agent automation is not a single app. It is a system made of three cooperating layers: orchestration, shared context, and operator visibility. Remove one layer and the whole thing becomes fragile.

Execution topology

Issue intake becomes a bounded run with memory, logs, and operator review

live routing model

Run queue

24 active

Context sync

97.8%

Operator reviews

3 pending

Orchestration layer

Symphony Workspace

Issues become bounded runs with explicit handoffs, evidence, and queue-level visibility instead of ad hoc agent sessions.

View Project

Shared memory layer

Second Brain

Humans and agents operate on the same durable memory base with explicit links, explainable retrieval, and context packet generation.

View Project

Operator layer

Agents Monitoring

A control-room console surfaces live sessions, task linkage, sync health, and the attention queue needed to operate real agents safely.

View Project

Design rules

Agents should run in bounded workflows, not open-ended chat loops.

Memory must be durable, inspectable, and shared across human and agent access paths.

Monitoring must show what is running, what is blocked, and what requires intervention.

Systems Proving The Stack

These projects show the operating model in pieces: orchestration, shared memory, monitoring, and the industrial reliability habits that keep automation usable outside a demo.

Workflow Orchestration

Symphony Workspace

Autonomous Issue Orchestration

An AI-native orchestration workspace that turns tracker issues into isolated implementation runs. Symphony lets teams manage work at the queue level while agents produce code, proof-of-work, and safe review handoffs.

Second Brain

One Durable Memory Shared by a Human and Two AI Assistants

A shared memory substrate where a human and two AI assistants — Claude Code and Hermes — operate on the same durable store through an MCP server. Postgres + pgvector with local Ollama embeddings, evidence-first ingest, typed relationships, explainable hybrid retrieval, graph-aware context packets, and automated loops that ingest, enrich, and resurface knowledge.

Ollama Embeddings (nomic-embed-text)

View Case Study

Operator Visibility

Agents Monitoring

Operator Console for Live Agent Systems

A control-room dashboard for watching real agents, live sessions, sync health, and task linkage on a single host. It is designed for operators who need to know what is running, what is blocked, and what requires intervention.

Industrial Edge Discovery

Diagnostics and Live System Documentation

A read-only platform for industrial edge support teams. It connects to remote devices, discovers running services and dependencies, generates live documentation, and shortens root-cause analysis with snapshot-based diagnostics.

Rule-Based Diagnostics

View Case Study

Industrial Telemetry

Industrial IoT Edge Platform

Operating a Multi-Site Data-Logging Fleet in Production

A production fleet of around 20 edge data-logging devices running across 10 industrial sites for different customers, on hardware from different brands. Machine and sensor data is acquired over Modbus, processed in containers at the edge, and streamed to cloud analytics — with one-click OTA updates and store-and-forward resilience.

Azure IoT Edge

Azure Pipelines (CI/CD + OTA)

AI Toolbox

Unified Interface for ML Pipelines

A web-based dashboard that consolidates disparate AI tools—text-to-speech, image generation, transcription—into a single, manageable interface. Eliminates CLI complexity for creative workflows.

FastAPI

Jinja2 Templates

HTMX

Celery-style async jobs

Kokoro TTS

View Case Study

Interactive Demos

Explore working prototypes of systems I've built. These dashboards demonstrate real-world interfaces with realistic data—but they're read-only simulations for demonstration purposes.

Read-only demonstrations. These interfaces showcase the actual design and functionality of production systems. All data is simulated—no backend services are running.

IoT Edge Monitor

Industrial Manufacturing

Real-time monitoring dashboard for industrial IoT edge devices. Track sensor data, device status, alerts, and data throughput across manufacturing floors.

PostFlow Analytics

Social Media Management

Analytics dashboard for social media scheduling platform. View post metrics, account status, queue depth, and engagement analytics in real-time.

AI Toolbox

Unified AI Tools Interface

Comprehensive AI tools dashboard for managing text-to-speech, image generation, transcription, and voice cloning jobs with real-time progress tracking.

Want to see the implementation details?

View Project Case Studies

Get in Touch

Available for contract work, full-time opportunities, and consulting engagements. I'm based in Zurich, Switzerland and work with clients globally.

Location

Zurich, Switzerland

Available for remote work and on-site engagements in Europe.

Central European Time (CET/CEST)

Response time is typically within 24 hours. For urgent inquiries, email is preferred.