Three generations of one machine
I build AI systems, measure whether they earn their keep, keep what does and replace what doesn't.
Lobster Actual and Hazel
Decommissioned both, verified with a cold boot
Inference and utility hub
Orchestrated Claude Code
Since March 2026 one machine has run three generations of AI systems. Each one was built to do real work and judged by whether it did. The rule is the same one I apply at GO2: governance is enforced in the system, not in a policy deck.
Lobster Actual and Hazel
Lobster Actual was a self-hosted 10-agent system built on OpenClaw. It ran as a hardened system service, with agents for orchestration, code review, research and intelligence, publishing and analysis.
- Models. A local-to-cloud fallback chain. Claude was structurally unreachable from sub-agent routing, so cost control never depended on an agent following a rule.
- Governance in three layers. Per-agent identity, a canonical rule set and a runtime contract.
- Observability. LobsterOps logged agent events, with Grafana and Prometheus for monitoring.
- Delivery. Autonomous review-then-merge pipelines across six product repos.
Hazel was a separate personal-life agent built on Hermes Agent and isolated from Lobster Actual by a strict credential wall. Two agents, one machine, zero shared credentials.
I wrote about the first build in I Built a $2,800 Mac mini Into a 24/7 Autonomous AI Agent. Here's Everything That Went Wrong (and Right). [1]
The decision
By July both agents had gone largely unused for weeks. I treated that as data. The question was whether the stack still earned its place, and the answer was no. I decommissioned both in September 2026 and verified the shutdown with a cold boot.
Most AI programs skip this step. Usage drops, the system keeps running and nobody owns the call to turn it off.
Retiring a system on evidence is part of governing it.
Inference and utility hub
The second build started from usage, not ambition.
- A hybrid local and cloud roster of 11 models served across my private network.
- An MLX server measured at about 2x Ollama throughput on the same 27B model: about 14.5 versus 7.7 tokens per second. [2]
- A phone-friendly chat interface and uptime monitoring.
- Encrypted offsite backups with a verified restore.
The standing rule: run it for a month and let real usage decide what gets added.
Orchestrated Claude Code
Generation 3 builds on Generation 2 instead of replacing it. The inference hub earned its place and stays as the base layer. Orchestration sits on top.
- An orchestrator agent runs role-scoped bots for engineering, experiments, ideas and design. Claude Code on the same machine does all the building.
- A model and effort routing rubric decides which model handles which work, with an escalation policy when a step fails.
- Credential and connector isolation is enforced in software, so work data and keys cannot reach agent sessions.
| Bot | Can read | Can write | Needs approval |
|---|---|---|---|
| Engineering | Product repos, open issues | Feature branches, pull requests | Merge to main |
| Experiments | Sandbox repo, public docs | Sandbox branches | Any new connector |
| Ideas | Notes, backlog | Backlog entries | None. No code access. |
| Design | Reference library, design canvas | Design files, design branches | Publishing outside the canvas |
| Orchestrator | Task queue, run logs | Task assignments | Model escalation, scope changes |
- Every change ships through a branch, a pull request and automated Claude review.
- 09:02:11OrchestratorAssigns the task to the Design bot.
Model and effort set by the routing rubric.2s - 09:02:13Claude CodeCreates a branch and builds.
branch design/reference-board6m 12s - 09:08:25Pull requestOpens with a summary of the change.4s
- 09:08:29Claude reviewAutomated review of the diff.
2 comments, 0 blocking. Both resolved.1m 08s - 09:09:37Human merge gateWaits for a person. Nothing merges without approval.
Approved. Merged to main.3m 41s
- A design-first pipeline: reference libraries, a design canvas, taste and component skills and an accessibility and UX audit.
What carries over
Three generations, one rule: keep what earns its place, replace what doesn't and put the rules inside the system. At GO2 that means a formal intake process for every AI tool and a policy that sets how much autonomy agents get. On my own hardware it means per-agent identity, credential walls and a review gate on every change.
The thinking behind both is in Agents Don't Need More Power. They Need Better Gates., Human in the Loop Is a Design Claim and The Rogue Agent Problem Is Already Here.