Skip to content

Topic

Durable Execution

A programming pattern for workflows that persist execution state so long-running, multi-step processes resume exactly where they left off after failures.

Current stories

buildOne report1 publisher

Omnara turns its remote-coding app into an Apache 2.0 control plane for durable agents

Omnara has replaced its remote-coding app with an Apache 2.0 control plane that keeps each agent's identity and history while worker machines come and go. Teams get one more self-hostable backend for agents run as durable services, though the only usage figures so far describe the retired app.

Publishers:runtimewire.com

Reality

Evidence45
Adoption
Not scored
Hype gap+5
Incentives60
Confidence50
productConfirmed5 publishers

Restate raises $20 million to take its durable execution runtime to San Francisco

Restate, the Berlin startup built by Apache Flink's creators, raised a $20 million Series A led by Singular, lifting its total funding to $27 million. The money pays for engineers and a San Francisco commercial hub to sell software that lets AI agents resume work after a crash.

Perspective Coverage

5 publishers
Builder
Builder 32%
Operator
Operator 20%
Investor
Investor 48%

Reality

Evidence64
Adoption42
Hype gap+22
Incentives58
Confidence68
buildOne report1 publisher

Pi 1.0 runs MCP tools inside a sandbox so their output stays out of context

Earendil's Pi 1.0 coding agent ships native MCP support after more than a year of criticising the protocol, running tool calls inside a WASM sandbox. The team credits a more stateless July 2026 spec and a sandbox it already needed for non-LLM models, so MCP support was a small addition.

Publishers:dev.to

Reality

Evidence40
Adoption35
Hype gap+20
Incentives40
Confidence40
buildOne report1 publisher

Lambda durable function replay turned a human-approved refund into an escalation

One developer's Lambda durable function executed 'escalate' after a human approved 'refund', because replay ran the uncheckpointed agent call a second time. Whatever a human signs off on has to be checkpointed with context.step so every replay returns the same value.

Publishers:dev.to

Reality

Evidence55
Adoption
Not scored
Hype gap+10
Incentives
Not scored
Confidence60
investConfirmed4 publishers

Lightspeed prices Temporal at 50 times revenue in a $550 million Series E

Temporal says its annualised revenue has passed $250 million and that its cloud processed more than 1.9 trillion actions in August. At the growth floors the company disclosed, volume is outrunning the revenue it produces.

Perspective Coverage

4 publishers
Builder
Builder 26%
Operator
Operator 23%
Investor
Investor 51%

Reality

Evidence55
Adoption70
Hype gap+20
Incentives70
Confidence60
buildOne report1 publisher

Hashing the arguments lets a reworded retry fire the destructive call twice

Across 34 runs on three agent frameworks, a recorder proxy shows that what the dedup key names decides whether a retried publish executes once or twice, and that surviving a SIGKILL is a separate question about where state is kept.

Publishers:dev.to

Reality

Evidence46
Adoption
Not scored
Hype gap+18
Incentives36
Confidence52
buildOne report1 publisher

The sub-second saga in this BIAN design tolerates 26.5 seconds before it compensates

A dual-orchestration pattern hands machine sagas to Temporal and human approvals to Flowable along BIAN service-domain lines. The retry policy in its Go sample shows what the machine side's latency budget really allows.

Publishers:dev.to

Reality

Evidence42
Adoption
Not scored
Hype gap+38
Incentives55
Confidence52
buildOne report1 publisher

Temporal's $550M round values durable execution at 2.5x its February price

Lightspeed co-led a $550 million Series E on September 14th, seven months after Temporal's $5 billion Series D. The operating metrics behind the markup are Temporal's own: 4,300 paying customers and 1.9 trillion billable actions in August.

Publishers:runtimewire.com

Reality

Evidence45
Adoption62
Hype gap+22
Incentives78
Confidence52
buildOne report1 publisher

A primary key and a lease column let Kestrel run durable workflows inside Postgres

Kestrel Workflows built durable execution out of Postgres row locks, constraints and leases, and published the schema that does it. The design keeps a second stateful system out of the path of every incident response.

Publishers:infoq.com

Reality

Evidence58
Adoption22
Hype gap+18
Incentives70
Confidence60
buildOne report1 publisher

Vercel puts agent-written TypeScript in QuickJS and makes host functions the real boundary

The Run SDK swaps eval's ambient authority for a worker-thread sandbox with narrow host functions and resumable interrupts. The replay semantics are the part to read twice.

Publishers:blog.vercel.com

Reality

Evidence45
Adoption22
Hype gap+18
Incentives78
Confidence50
buildOne report1 publisher

Cloudflare CASB now revokes the risky share before an admin opens the finding

Cloudflare's new remediation policies match a finding against a config, then run the revoke or the webhook on Workflows. The stated outcome is minutes, and it holds while the SaaS API is not throttling you.

Reality

Evidence38
Adoption15
Hype gap+20
Incentives85
Confidence45
buildOne report1 publisher

Spring's @Scheduled swallows the exception and reports success to the scheduler

A nightly job that catches its own failure and returns cleanly is invisible to every dashboard you can bolt on afterwards, which is why the durable execution argument lands on the event log rather than the feature list.

Publishers:dev.to

Reality

Evidence38
Adoption
Not scored
Hype gap+30
Incentives50
Confidence55
productOne report1 publisher

The agent bottleneck is a workflow engine your team wrote by accident

A devops.com argument worth taking seriously: the hard part of production agents is what happens after step one, and the fix is orchestration engineering, not prompting.

Publishers:devops.com

Reality

Evidence28
Adoption
Not scored
Hype gap+30
Incentives62
Confidence42
buildOne report1 publisher

Four control planes, one Postgres: a team's case against polyglot persistence

A dev.to account of retiring Mongo, Redis, Elasticsearch and a vector store makes its strongest argument on consistency rather than cost, and it arrives without a single figure.

Publishers:dev.to

Reality

Evidence24
Adoption34
Hype gap+46
Incentives62
Confidence30
productOne report1 publisher

A retry that sends the email twice is a correctness bug, not a slow job

A systems newsletter makes the case for durable execution in agent workflows. The consequence for operators is narrower and harder: side effects have to be safe to replay.

Publishers:newsletter.systemdesign.one

Reality

Evidence26
Adoption
Not scored
Hype gap+22
Incentives74
Confidence34
productOne report1 publisher

Agents Are Not Microservices With an LLM Attached, and the Retrofit Never Arrives

A devops.com essay argues durability, per-step identity and decision-level observability belong in the agent runtime, not layered on later. The failure economics support the claim.

Publishers:devops.com

Reality

Evidence24
Adoption
Not scored
Hype gap+34
Incentives55
Confidence38
buildOne report1 publisher

Temporal talks $12B six months after $5B: durable execution is now agent infrastructure

A reported $500 million round at a pre-money floor of at least $12 billion reprices a two-decade systems bet. Teams already running workflows on it should treat this as a procurement event.

Publishers:runtimewire.com

Reality

Evidence55
Adoption66
Hype gap+27
Incentives70
Confidence54

Earlier coverage