Technical Journal

Engineering Insights

Notes on robotics research, AI systems architecture, and the lessons learned from building physical products.

Articles
2026-09-06

Nesting Hermes Inside Grok Bot: Same Box, Two Stacks, One Pipeline

Why I stopped running Hermes on a VPS and nested it inside the Chief of Staff's box — cheap long-haul capacity under expensive orchestration.

AIAgentsGrok BotHermesOllamaArchitectureToken Economics
Read Entry
2026-08-27

From a Wheelbarrow to an Electric Car

Three generations of a personal AI assistant I actually use: OpenClaw on a VPS, Hermes on Telegram, and Grok Bot with a desktop. Not a setup log — the work it carries.

AIAgentsOpenClawHermesGrok BotObsidianProductivity
Read Entry
2026-08-25

Certifying an Autonomous Forklift in Germany: ISO 13849, ISO 3691-4, and the Road to CE

The full certification procedure for factory-floor autonomous forklifts in Germany — from ISO 12100 risk assessment through ISO 13849-1 performance levels and ISO 3691-4 safety zones, to the safety hardware, the F-CPU software architecture, and the CE marking roadmap.

ISO 13849ISO 3691-4Functional SafetyAGVForkliftCE MarkingMachinery Regulation
Read Entry
2026-08-25

Certifying an Autonomous Forklift in the United States: ANSI B56.5, R15.08 & OSHA

The US has no CE mark. Instead, a stack of OSHA regulation, ANSI consensus standards, and UL/NRTL listing governs autonomous forklifts — ANSI/ITSDF B56.5, ANSI/RIA R15.08, OSHA 1910.178, NFPA 79, and UL 3100. Here is the full compliance roadmap.

ANSI B56.5R15.08OSHAFunctional SafetyAGVForkliftULNRTL
Read Entry
2026-08-25

EU vs US: Two Ways to Certify an Autonomous Forklift

The EU and the US certify the same autonomous forklift through fundamentally different systems — a preventive, harmonized CE-marking regime versus a decentralized, liability-driven OSHA/NRTL landscape. A side-by-side comparison of the legal basis, conformity assessment, functional safety, electrical standards, cost, and market-entry strategy.

Functional SafetyAGVForkliftCE MarkingOSHAISO 3691-4R15.08UL
Read Entry
2026-08-24

What is VDA 5050? The Universal Language of AGV Fleets

Vendor lock-in is the core problem VDA 5050 solves. A look at the centralized master-control architecture, MQTT as transport, and the five message topics that let heterogeneous AGVs from different vendors work as one fleet.

VDA 5050AGVAMRMQTTFleet ManagementIntralogistics
Read Entry
2026-08-24

MQTT Topics & Message Schemas: The VDA 5050 Wire Protocol

Field-by-field through the VDA 5050 wire protocol: the MQTT topic hierarchy, QoS and retain flags, the common message header, and the five schemas that define every message on the wire.

VDA 5050MQTTAGVAMRFleet ManagementProtocol
Read Entry
2026-08-24

Order Lifecycle & State Machine: How an AGV Executes an Order

The transport order lifecycle in VDA 5050: orderId and orderUpdateId, nodes and edges, the released base vs horizon split, the order acceptance procedure, and how a vehicle reports progress through nodeStates and edgeStates.

VDA 5050AGVAMRState MachineFleet Management
Read Entry
2026-08-24

Instant Actions & the Connection Watchdog: Safety in a Networked Fleet

The instantActions topic and its eight standard actions, blockingType semantics, the actionStates lifecycle, and the connection topic's last-will pattern that turns a network drop into a controlled stop.

VDA 5050AGVAMRSafetyFleet Management
Read Entry
2026-08-24

Fleet Integration, Traffic & Deadlock: VDA 5050 in Production

The final piece: a fleet manager architecture, traffic control with a floor ledger, wait-die deadlock handling, the factsheet as a capability contract, and how VDA 5050 compares to MassRobotics AMR interoperability.

VDA 5050AGVAMRTraffic ControlDeadlockFleet Management
Read Entry
2026-08-01

What Graph Engineering Actually Looks Like in Practice

A practitioner's account of building a persistent agent system on Anthropic's three-layer architecture — harness, loops, graph — and what broke before it worked.

AI AgentsGraph EngineeringHarness EngineeringLoop EngineeringAnthropicKnowledge GraphAgent Architecture
Read Entry
2026-07-17

Why I Built a ROS 2 SIL Platform from Scratch

Robotics teams keep hitting the same integration problems: sim passes but hardware fails, PRs break on merge, tests are flaky. Here's how a Software-in-the-Loop test platform solves them at the root.

ROS 2CI/CDRoboticsSILIntegration
Read Entry
2026-07-17

Architecture: HAL Boundary, Sim Time, and 6 ADRs

The architecture behind a ROS 2 + MuJoCo SIL platform: a single control loop, a hardware abstraction layer for sim-to-real, sim time for deterministic CI, and six documented architecture decisions.

ROS 2ArchitectureSim-to-RealHALADRMuJoCo
Read Entry
2026-07-17

CI/CD for Robotics: GitHub Actions, Merge Queue, Docker

The full anatomy of a robotics CI pipeline: GitHub Actions with container jobs, caching strategy for sub-10-minute builds, native merge queue setup (including the 422-to-org war story), multi-stage Docker, and GHCR release.

CI/CDGitHub ActionsDockerMerge QueueGHCRRobotics
Read Entry
2026-07-17

Testing Robotics Code: 45 Tests Across Four Layers

A four-layer test pyramid for robotics: linters for hygiene, pure-math gtests because the logic is deliberately separated from nodes, a smoke test that caught a silent QoS mismatch, and a sim-time SIL scenario that caught a physics bug no unit test could see.

TestingROS 2gtestSILSim TimeQoS
Read Entry
2026-07-17

6 Real Bugs That SIL Testing Caught

Six real bugs from a robotics CI/CD platform: a silent QoS mismatch, a physics-induced false obstacle detection, a Dockerfile that hid a missing package, and more. Each with symptom, diagnosis, fix, and lesson.

DebuggingROS 2TestingSILBugsRobotics
Read Entry
2026-07-17

Release Engineering: From Tag to Traceable Image

How a robotics CI/CD platform goes from git tag to versioned Docker image on GHCR — with a fully traceable chain from digest to commit, and a lesson on why you should never re-point a published tag.

Release EngineeringGHCRDockerSemverCI/CDRobotics
Read Entry
2026-07-17

From Integration Pain to Production Pipeline

What building a ROS 2 SIL CI/CD platform from scratch taught me about integration engineering, scope discipline, and why the pipeline — not the robot — is what keeps hardware safe and teams fast.

RoboticsCI/CDSILLessonsIntegration
Read Entry
2026-05-13

Meine A2.1 Notizen: Der komplette Grammar & Vocabulary Guide

My complete A2.1 German study notes from Momente A2.1 — all grammar tables, vocabulary lists, and the patterns that finally made things click.

GermanDeutschA2Language LearningGrammarVocabulary
Read Entry
2026-04-29

Building an AI-Powered Personal Knowledge Pipeline: Obsidian + NotebookLM + OpenClaw

How I automated my learning workflow so my AI assistant handles the boring parts — scheduling, summarizing, and reminding.

AIAutomationObsidianProductivityWorkflow
Read Entry
2025-03-01

50 Lessons from Building a Humanoid VLA System

Engineering lessons learned while building a Vision-Language-Action system for the Unitree G1 humanoid robot in MuJoCo.

RoboticsMuJoCoImitation LearningROS 2
Read Entry

hello-world-meet-rookie

Read Entry
Index: 0x32 entries
EOF