Independent deployments will determine whether Machine0’s CLI- and MCP-managed persistent CPU and GPU VMs provide reliable, economical infrastructure for multi-day agent workloads.

state: expiredheat: lowuncertainty: highknownscott: lowlong-running-agents agent-infrastructure ai-infrastructureBarnabyMachine0

What is this?

Machine0 is presented in its Launch HN and YC launch materials as a CLI for provisioning persistent CPU and GPU cloud VMs for agents, with per-minute billing advertised from $0.013/hour; the supplied material also associates Barnaby with the launch, but does not establish his role. The broader snippets describe persistent environments, on-demand GPUs, and MCP infrastructure as options for long-running agent execution, but they come largely from infrastructure vendors and do not provide independent deployment evidence about Machine0’s reliability or economics. Whether Machine0 performs economically and reliably over multi-day workloads therefore remains unverified here.

Why it matters to Scott

Machine0 repeats the persistent Linux-workshop substrate already argued in “Give the Agent a Workshop” and “long-running-agents,” while the radar already tracks substantially similar validation questions in “Fly.io will turn its bet on computers for AI agents…” and “AWS Bedrock AgentCore runtime instances…”. It could eventually affect Scott’s infrastructure choices, but without independent reliability or cost evidence this launch is only another vendor example of an established pattern.

ip:framework.long-running-agentsip:source.give-the-agent-a-workshop-ebookip:concept.goal-world-isolationdev:concept.resumable-agent-job-control-planeradar:fly-agent-computers-infrastructureradar:aws-agentcore-persistent-runtime-adoptionradar:concept.agent-infrastructureradar:concept.long-running-agents

queries asked of Scott's wikis

persistent compute for long-running coding agents
agent sandbox persistence versus ephemeral execution
CLI-managed agent infrastructure
MCP-managed remote execution environments
economics of CPU and GPU agent workloads
reliability patterns for multi-day agent runs

Measured heat

no measured readings yet — the hourly heat pass fills this in

How the heat travelled

no chain yet — the hourly chain pass fills this in

Evidence (3) — ⭐ canonical anchor

source	object	author	score	comments
🟧 hn	Launch HN: machine0 (YC S26) – Persistent CPU and GPU VMs from the CLI	bwm	77	42
🟧 echo.other ⭐	YC’s launch page says: “machine0 is a CLI for agent compute” that gives agents persistent cloud VMs, billed by the minute, from $0.013/hour	Barnaby Malet	—	—
🟧 hn	Show HN: Yado – Claude Code and Codex on Your iPhone	mehdigtb	1	1

Interpretation history

2026-08-21T15:36:26Z
No independent Machine0 deployment, reliability, or cost evidence emerged during the observation window; the launch has faded without advancing beyond an unvalidated example of an established persistent-agent infrastructure pattern.
2026-08-19T14:38:09Z
Yado independently demonstrates the broader product pattern of persistent per-user VMs for remote coding agents, but it uses Fly.io and supplies no deployment, reliability, or economic evidence for Machine0 itself. The case therefore remains a vendor claim awaiting independent Machine0 workloads rather than corroboration of its specific hypothesis.
2026-08-19T14:23:35Z
evidence attached: hn.story.49361945 — A usable first-party-like implementation shows persistent Fly.io machines and stream-json relays supporting unattended mobile Claude Code and Codex sessions, materially contextualizing persistent agent runtimes.
2026-08-19T05:26:17Z
The refreshed discussion remains repetitive product questioning and pricing skepticism, with no independent deployment, reliability, or multi-day cost evidence. Machine0 remains an unvalidated vendor implementation of an established persistent-agent infrastructure pattern.
2026-08-19T00:26:57Z
Refreshed comments still consist of pricing skepticism and questions about provider, differentiation, and suspend/resume economics. No independent deployment or operational evidence changes the case from an unvalidated vendor implementation of an established pattern.
2026-08-18T21:40:45Z
Refreshed discussion remains focused on basic differentiation, pricing and suspend/resume questions rather than independent deployments. The case is still an unvalidated vendor implementation of an established persistent-agent infrastructure pattern.
2026-08-18T17:33:08Z
The small engagement increase adds no independent deployment, reliability, or cost evidence; this remains an unvalidated vendor example of an already established persistent-agent infrastructure pattern.
2026-08-18T17:31:20Z
grounded: known/low — Machine0 repeats the persistent Linux-workshop substrate already argued in “Give the Agent a Workshop” and “long-running-agents,” while the radar already tracks
2026-08-18T17:27:46Z
origin walked (codex/luna, conf 0.96): anchor hn.story.49348136 -> echo.other.68dfc08a4c by Barnaby Malet
2026-08-18T17:26:23Z
case created — The founder’s launch provides a concrete persistent-compute service purpose-built for long-horizon agents, distinct from existing durable-runtime and distributed-job cases.

Decision trace

08-22 01:36expireNo independent Machine0 deployment, reliability, or cost evidence emerged during the observation window; the launch has faded without advancing beyond an unvalidated example of an established persiste
08-22 01:36alert_silentThe only new signal is elapsed time without substantive evidence, which neither changes Scott’s decisions nor warrants attention outside routine rediscovery if deployments later appear.
08-22 01:36alert_routeThe only new signal is elapsed time without substantive evidence, which neither changes Scott’s decisions nor warrants attention outside routine rediscovery if deployments later appear.
08-20 00:38repriceYado independently demonstrates the broader product pattern of persistent per-user VMs for remote coding agents, but it uses Fly.io and supplies no deployment, reliability, or economic evidence for Ma
08-20 00:38alert_silentThe adjacent Yado implementation reinforces an already-established infrastructure pattern but does not change the evidence on Machine0’s reliability or multi-day economics, so routine review is suffic
08-20 00:38alert_routeThe adjacent Yado implementation reinforces an already-established infrastructure pattern but does not change the evidence on Machine0’s reliability or multi-day economics, so routine review is suffic
08-20 00:24alert_silentYado is a small first-party product launch packaging Claude Code and Codex with a per-user Fly.io machine and an iPhone interface. It illustrates demand for persistent remote agent execution but provi
08-20 00:24alert_routeYado is a small first-party product launch packaging Claude Code and Codex with a per-user Fly.io machine and an iPhone interface. It illustrates demand for persistent remote agent execution but provi
08-20 00:23attachA usable first-party-like implementation shows persistent Fly.io machines and stream-json relays supporting unattended mobile Claude Code and Codex sessions, materially contextualizing persistent agen
08-20 00:22propose_attachA usable first-party-like implementation shows persistent Fly.io machines and stream-json relays supporting unattended mobile Claude Code and Codex sessions, materially contextualizing persistent agen
08-19 19:21sensor_dirtyengagement_update
08-19 17:21sensor_dirtyengagement_update
08-19 15:26repriceThe refreshed discussion remains repetitive product questioning and pricing skepticism, with no independent deployment, reliability, or multi-day cost evidence. Machine0 remains an unvalidated vendor
08-19 15:26alert_silentNo consequential new fact has emerged beyond refreshed discussion and engagement; independent operational validation is still absent, so this can wait for routine review.
08-19 15:26alert_routeNo consequential new fact has emerged beyond refreshed discussion and engagement; independent operational validation is still absent, so this can wait for routine review.
08-19 14:21sensor_dirtycomment_update
08-19 12:21sensor_dirtyengagement_update
08-19 11:21sensor_dirtycomment_update
08-19 10:26repriceRefreshed comments still consist of pricing skepticism and questions about provider, differentiation, and suspend/resume economics. No independent deployment or operational evidence changes the case f
08-19 10:26alert_silentThe discussion adds no verified reliability, cost, or multi-day workload results, so it can wait for routine review.
08-19 10:26alert_routeThe discussion adds no verified reliability, cost, or multi-day workload results, so it can wait for routine review.
08-19 09:21sensor_dirtycomment_update
08-19 08:21sensor_dirtyengagement_update
08-19 07:40repriceRefreshed discussion remains focused on basic differentiation, pricing and suspend/resume questions rather than independent deployments. The case is still an unvalidated vendor implementation of an es
08-19 07:40alert_silentThe new comments add skepticism and product questions but no verified reliability, economics or deployment evidence, so the delta can wait for normal review.
08-19 07:40alert_routeThe new comments add skepticism and product questions but no verified reliability, economics or deployment evidence, so the delta can wait for normal review.
08-19 06:21sensor_dirtyengagement_update
08-19 05:21sensor_dirtycomment_update
08-19 04:21sensor_dirtycomment_update
08-19 03:33repriceThe small engagement increase adds no independent deployment, reliability, or cost evidence; this remains an unvalidated vendor example of an already established persistent-agent infrastructure patter
08-19 03:33alert_silentOnly repetitive launch amplification has arrived, so there is no consequential new delta that Scott needs before a normal briefing.
08-19 03:33alert_routeOnly repetitive launch amplification has arrived, so there is no consequential new delta that Scott needs before a normal briefing.
08-19 03:31alert_silentMachine0’s launch is an established vendor event, but it adds another CLI/MCP-managed persistent-VM offering to an already tracked infrastructure pattern. Its uptime, reliability, security and workloa
08-19 03:31alert_routeMachine0’s launch is an established vendor event, but it adds another CLI/MCP-managed persistent-VM offering to an already tracked infrastructure pattern. Its uptime, reliability, security and workloa
08-19 03:31groundMachine0 repeats the persistent Linux-workshop substrate already argued in “Give the Agent a Workshop” and “long-running-agents,” while the radar already tracks substantially similar validation questi
08-19 03:27promote_anchororigin walk conf 0.96
08-19 03:26createThe founder’s launch provides a concrete persistent-compute service purpose-built for long-horizon agents, distinct from existing durable-runtime and distributed-job cases.