Temporal Technologies has raised $550 million at a $12.55 billion valuation, and the pitch that closed the round is unglamorous by design: AI agents keep dying halfway through their jobs, and somebody has to sell the software that picks them back up.
The Series E, announced September 14, was led by Lightspeed Venture Partners with Wellington Management, Growth Equity at Goldman Sachs Alternatives, and Tiger Global as co-leads. T. Rowe Price and SV Angel joined as new backers; a16z, Sequoia, Index, GIC, Sapphire Ventures, and Amplify returned. Total capital raised now stands at roughly $1.2 billion, and the valuation is more than double the $5 billion the Bellevue, Washington company commanded at its a16z-led $300 million Series D in February. Go back one more step and the repricing is starker still: Temporal's Series C in March 2025 valued it at $1.72 billion. That is a 7.3x markup in 18 months for a company that sells workflow orchestration.
The numbers Temporal disclosed alongside the round are the reason investors are willing to pay up. Annualized revenue run rate recently passed $250 million, growing more than 200% year over year. Net dollar retention has held above 200% since February, meaning the average customer from a year ago is now spending more than twice as much. Temporal Cloud processed 1.9 trillion billable actions in August, up more than 350% year over year. Open-source installs crossed 43 million, up 134% since January. Paying customers number more than 4,300, up 139%, and the roster includes OpenAI, Snap, NVIDIA, Netflix, JPMorgan Chase, Anysphere (the maker of Cursor), Lovable, Scale AI, Salesforce, Shopify, and DoorDash. Headcount has doubled to 570 in a year.
"As agents take on more critical work across more systems, every additional step creates another place to fail. In production, that work has to survive those failures and finish reliably," said Samar Abbas, Temporal's co-founder and CEO, in the announcement. "Temporal was built for this problem. Durable Execution is becoming the standard for reliable applications at scale, and this investment reflects the conviction that much of the next generation of software will be built on Temporal."
What Temporal actually sells is a runtime that checkpoints the state of a running program at every step. When a process crashes, an API times out, or a server disappears, the workflow resumes from the last recorded step instead of starting over. The company calls this Durable Execution. It was originally sold to teams running payments, onboarding, and order fulfillment across distributed systems, and it is not new technology: Abbas and CTO Maxim Fateev built Amazon's Simple Workflow Service together, then Microsoft's Durable Task Framework and Uber's Cadence, before founding Temporal in 2019.
What changed is the workload. An AI agent that plans a task, calls a dozen tools, waits on a model response, and writes results to three systems is exactly the kind of long-running, multi-step, failure-prone process Temporal was built for, except that now it runs for hours or days rather than seconds. Temporal's own blog put it plainly: a year ago its customers were talking about payments; now they are asking to run agents "for days, weeks, or months." OpenAI's use of the platform has grown 60-fold in under a year, according to the company, and Snap moves 414 million Stories a day through it.
"Every team building on AI hits the same wall: the demo is easy, production is hard, because the systems around the models can't handle real-world execution," said Anoushka Vaswani, partner at Lightspeed. "Most of the market solves that by locking teams into a proprietary stack. Temporal is the open, pluggable foundation that runs their agents and the systems they already depend on, and it's becoming one of the most important pieces of infrastructure in the agentic era."
Venkat Venkataramani, OpenAI's VP of Infrastructure, was more specific about why his team bought in: "Durable Execution is more than ever a core requirement for modern AI systems, and Temporal offers a compelling platform to help build it in from the start. This is one of the main reasons why we invested in building a durable orchestration framework powered by Temporal at OpenAI."
Why It Matters
Start with the multiple. At $12.55 billion on a run rate just over $250 million, Temporal is priced at roughly 50 times current annualized revenue. That is not a software multiple; it is a bet that 200% growth and 200% net retention persist long enough for the revenue to catch up to the price. If growth holds for two more years, the company would be at roughly $2.25 billion in ARR and the round looks cheap. If the agentic-AI buildout slows, or if the hyperscalers decide durable orchestration should be a free feature of their agent platforms, the math gets uncomfortable fast. Tech Funding News noted the valuation has now doubled twice in a single year and asked whether that reflects a genuine shift in what durable execution is worth or "a market getting ahead of itself." Both can be true.
The second point is where the money in AI is actually landing. Temporal does not train models, host inference, or ship an agent product. It sells the state machine underneath other people's agents, and it bills per action, which means every additional tool call an agent makes is revenue. The 350% jump in billable actions against a 200% jump in revenue suggests the pricing per action is falling even as volume explodes, which is normal for infrastructure at scale but worth watching. The lesson for the rest of the market is that the least visible layer of the agent stack is producing some of the most defensible numbers.
Third, the competition is real but small. Inngest has raised about $34 million in total. Trigger.dev closed a $16 million Series A in December. Restate, founded by Apache Flink alumni, has raised roughly $7 million. DBOS, backed by Michael Stonebraker and Matei Zaharia, is still at seed stage. None are within two orders of magnitude of Temporal's funding, but the fact that a half-dozen well-credentialed teams are building in the category tells you investors see a market, not a monopoly. The more immediate threat is the open-source core itself: 43 million installs is a moat when they convert to Temporal Cloud and a liability when a large customer decides to self-host.
What to Watch
The next data point is whether Temporal's per-action pricing holds as its largest customers scale. OpenAI's 60x growth in usage is the headline, but the company that grows 60x on your platform is also the company best positioned to negotiate the rate down or build its own. Watch for Temporal's next disclosed run-rate figure, likely around the company's Replay conference cycle, and whether net retention stays above 200% once the early-adopter cohort matures. Watch, too, for AWS, Google, or Microsoft to bundle durable orchestration into their agent frameworks; Microsoft already ships Azure Durable Functions on a framework Abbas co-created. And with $1.2 billion raised and a $12.55 billion price tag, the question of an IPO window in 2027 is no longer hypothetical.
"Every team building on AI hits the same wall: the demo is easy, production is hard, because the systems around the models can't handle real-world execution."— Anoushka Vaswani, Partner, Lightspeed Venture Partners