Temporal Technologies

What can we say? It's a banger.

Demos make agents look easy. Production is where you find out what the demo never had to keep alive. This Wed on Vibe Check, Melanie Warrick is exploring AI memory and state with two people who work on this for a living: Christopher Gambill, principal architect at Gambill Data, and Scott Haines, Staff Developer Advocate at Databricks. Chris and Scott will share what they've been building and how they think about putting AI in production, especially the parts that are hard to fake: managing state and memory in agents that run for a long time. They'll get into what's even worth keeping once persistent memory grows unbounded, and how much an agent should carry from one session to the next. Join the conversation this week!

Shy Ruparel bought a Japanese collectible and ran into an immediate problem: he could not read the manual. So he built Toku Solutions, a project that turns Japanese toy manuals into translated, editable sites. How he used Temporal to run the OCR, translation, image rendering, and AI cleanup without having to start the whole pipeline over when something failed: https://lnkd.in/d6jYbaEc

⚽ We're hosting a meetup at Selhurst Park next month ⚽ Not only will you sample the amazing atmosphere of a South London icon, you'll get to hear from three speakers who'll give you their engineering insights into how you can make your code run more durably at scale. Whether you're building AI agents, backend services, or long-running business processes, you'll leave with practical ideas that make applications more reliable, resilient, and easier to operate in production. More details and registration: https://luma.com/sh4j0dge

Hey, NYC 👋 Have plans this Thursday night? Cancel them immediately. Instead, join us in partnership with Box, Anthropic, and Auth0 for AI Hack Night NYC. There will be food, demos on the newest models and tools for building with agents, and plenty of time to work on whatever you can think of. So bring your laptop, your brain, and your appetite. They’ll all be used. Spots are limited. Register: https://luma.com/lxakd22n

❤️💙 It's time. Meet Crystal Palace Football Club's first kit with Temporal as the front-of-shirt sponsor. To mark the beginning of our partnership, we created a short film inspired by 50 years of the iconic sash and the people who've carried it forward. Thanks to our partners Milk & Honey Productions, Mad Cow Films, UNIT, and Playfly Sports for helping bring it to life. Learn more about our partnership in the comments.

Temporal’s LangGraph and LangSmith integrations are now in Public Preview. If you’re building agents with LangGraph, you can now run them on Temporal without rewriting your codebase. Keep the framework you already use and add the durability your agents need in production. All the details: https://lnkd.in/dYGpdUG3

Nehal Agrawal, GTM engineer at Invigilo AI, joins Melanie Warrick and Melissa Herrera on Vibe Check this Wednesday at 11am PT to talk about autonomous agents in go-to-market. How much of the agentic GTM stack is actually autonomous, where is it pulling real weight, and where is it overhyped? Nehal will walk through a real outreach workflow she built that hit a 60% reply rate, end-to-end on screen, plus an honest read on where autonomy still breaks and where a human stays in the loop.

Just like fingerprints, irises, snowflakes, and burrito bowl orders, no two bookings at Shree Maruti (SMILe) are alike. And with each unique booking, the platform has to pick the right path every time–and fast. Before using Temporal, that was tricky. Now, every booking runs as a single Temporal Workflow from start to finish. The platform chooses a path, hands off to delivery partners, retries whatever fails, and never loses track of where things stand. Today, for SMILe’s infinite possibilities of bookings, there’s a single system for every kind of delivery, while every booking is traceable end to end. Read the full story: https://lnkd.in/dfSQf3gD

What changes when OpenBox AI's runtime governance becomes part of a Temporal Technologies workflow? Temporal already helps enterprises keep long-running workflows reliable across failures, restarts, and extended execution periods. With the OpenBox integration, teams can now govern what AI agents do inside those workflows before actions are executed. Temporal builders can now: • Enforce policies before workflows, activities, and signals run • Route sensitive actions for human approval • Keep approvals intact across workflow restarts • Generate cryptographically signed records for governed decisions • Add governance without rebuilding existing Temporal workflows Durable execution keeps the workflow running. Runtime governance helps ensure the agent only takes actions that are authorized, controlled, and auditable. For enterprises deploying autonomous agents across business systems, reliability is only the foundation. The next requirement is control: ensuring every action is authorized before execution and verifiable afterward.