Temporal Technologies

Shy Ruparel, our senior developer advocate, created a conference badge and made crashing your Worker a game mechanic.

More specifically, he put Temporal's Rust SDK 1.0 on 10 conference badges and let people play a trivia game on them. Hold LEFT + RIGHT and a badge "crashes." Its question gets retried on another badge, and the Workflow still knows everyone's score.

It took four small patches to get the real SDK running on an ESP32-S3 and one "completely reasonable" battery optimization to make the whole booth think the demo was broken.

Read the whole story: https://pulse.ly/tw51mrqif3

Well, this is fun: Temporal just placed number 1 (!) on the GeekWire 200. 🎉

Turns out keeping AI agents running when things get messy is a pretty big deal. We’re proud to be building that foundation from Bellevue (with amazing teammates all over the world).

See the list:

There’s a new No. 1 on the GeekWire 200 as AI and hardware reshape our Pacific Northwest startup index Temporal, whose software keeps AI agents running reliably, is the new No. 1 on the GeekWire 200. Companies building AI software and advanced machinery for space, energy and defense lead the Fall 2026 update of our ranking of the Pacific Northwest's top privately held tech startups.

Temporal’s Rust SDK is now generally available.

Rust developers can now write Workflows and Activities, run Workers on Tokio, and connect to Temporal Cloud or a self-hosted Temporal Service through native Rust APIs.

Temporal records Workflow progress so ex*****on can continue after a Worker stops. Running Workflows can wait on durable timers, receive Signals and Updates, and answer Queries. The SDK also includes Child Workflows, Continue-As-New, retries, cancellation, and timeouts.

Workflow state lives in Rust structs. Workflows and Activities are async functions. Procedural macros define them, and typed method references let Rust check inputs and results.

Quickstart: https://pulse.ly/3x5hlvgkzp

API documentation: https://pulse.ly/ycaqiuw8yo

GitHub: https://pulse.ly/whocwrsxjb

Try it and share your feedback in -sdk in the Temporal Community Slack: https://pulse.ly/75n2pkfglf

Experts from Oso have joined Temporal to accelerate our work on AI agent security. Welcome to the team!

As agents move money and call other agents, access decisions have to hold at every handoff, even through crashes and retries. Read more: https://pulse.ly/cskg01oewn

Serverless Workers for Google Cloud Run are now in public preview.

Run your Temporal Workflows and Activities on Cloud Run Worker Pools and let Temporal handle the scaling. Based on workflow volume, Temporal automatically scales your Workers up as work arrives and scales to zero when appropriate.

Get started: https://pulse.ly/lmhbwnunv7

Heading to MongoDB.local NYC tomorrow?

Catch our CTO, Maxim Fateev, after the opening keynote on a panel discussion with OpenRouter and Cognition on what it takes to run AI agents reliably in production.

And, hey: visit the Temporal booth to meet the team and participate in demos!

Full agenda: https://pulse.ly/0xbqawmxd1

Stuck waiting on a UI roadmap? One Temporal customer built their own Chrome extension instead.

Join us to see how they tailored Temporal's Web UI to their exact needs, and you'll walk away with the extension's architecture and a sample repo to start building your own.

Register: https://pulse.ly/bpwwwupue8

What if an AI agent could become durable without changing the agent code itself?

In this guest post, Greg Haskins, co-founder and CTO at Manetu, explains how AgentVisor maps Agent Protocol Threads to Temporal Workflows so agent state, checkpoints, and long-running sessions can survive crashes and resume elsewhere.

Read more: https://pulse.ly/c6aaikczhs

AI agent workloads are unpredictable. They can experience sudden bursts of activity, sit idle waiting for human input, or run for days.

Today, we're introducing the prerelease of Amazon Bedrock AgentCore runtime as a compute provider for Temporal Serverless Workers.

⚡ Scale Worker capacity up or down based on demand
🔄 Recover from failures without losing Workflow progress
🛠️ Deploy and operate agents without managing an always-on Worker fleet

See how it works and explore the reference architecture: https://pulse.ly/6j6tq5sqof

Coding agents can write a lot of code. Getting them to write correct Temporal Applications takes more than a good prompt.

Mason Egger shares the setup he uses to help agents build production-ready Temporal code, including the Temporal Developer Skill, SDK samples, the Knowledge Base MCP, and guardrails like testing and type checking.

Read more: https://pulse.ly/wsqqne4ck0