- Poetic raises $50M Series A at a $500M valuation led by Kleiner Perkins, with OpenAI, Founders Fund, and First Harmonic also investing
- The San Francisco startup hit an eight-figure run rate and turned profitable in 2025 with just four employees, automating fraud and compliance workflows for SoFi, AIG, and Chime.
- Unlike conventional AI agents, Poetic uses a proprietary programming language to turn business processes into deterministic, repeatable code.
Most enterprise AI pilots die quietly. They look impressive in a demo, survive a controlled test, and then stall the moment they hit a real compliance workflow, a fraud investigation queue, or a multi-hour insurance review.
Poetic, a San Francisco startup formerly known as Forge, thinks it has cracked the problem, and a $50M Series A suggests some of the sharpest investors in the business agree. The round was led by Kleiner Perkins at a $500M valuation, with participation from OpenAI, Founders Fund, First Harmonic, and Genius Ventures.
It is not Kleiner Perkins’ first bet on Poetic: the firm also led the company’s seed round, and partner Leigh Marie Braswell personally backed the company before it had a product to show. That kind of conviction, doubling down at a ten-figure valuation, says something about what Kleiner Perkins has seen in production.
The opportunity is large and largely unlocked
Global financial services firms spend billions each year on manual operational processes, including fraud investigations, transaction monitoring, compliance checks, insurance reviews, that existing software and AI tools have failed to reliably automate. The workflows are too complex, too full of unwritten rules, and too consequential to hand off to a system that might be right 90% of the time.
Poetic was founded in 2023 by Markie Wagner, a Thiel Fellow and Stanford computer science graduate who previously worked as an ML engineer at Google and Waymo before founding Delphi Labs, an ML consultancy that deployed AI for some of the world’s largest organisations. That background, seeing AI fail in production, over and over, across different industries, shaped what she built next.
The company’s core idea is a departure from the autonomous agent model that most enterprise AI startups are chasing.
Rather than deploying a system that reasons freely through a workflow, Poetic uses a proprietary programming language that lets operations teams describe complex processes in natural language, then converts that logic into deterministic, repeatable execution. It learns like AI, but runs like code. The result, the company says, is automation that can handle multi-hour processes running thousands of times a day with near-zero tolerance for error.
The numbers in production are striking. At SoFi, Poetic reached 99%+ quality executing fraud investigations end-to-end in five weeks. AIG and Chime are also named customers.
The company says it has generated tens of millions of dollars in savings for Fortune 500 clients and has converted every pilot it has run into a production deployment.
One of the most competitive corners of enterprise
Harvey, which recently raised $300M at a $5B valuation, is targeting complex legal workflows. Abridge has raised $300M to automate clinical documentation in healthcare.
Most competitors, however, are building on top of general-purpose LLMs. which Poetic argues is the root cause of the reliability problem. Its language and deterministic execution model use significantly fewer tokens, which the company claims makes it cheaper and more controllable at scale.
“What Poetic has built is genuinely different — a platform that can execute the complex, high-stakes processes that large enterprises actually run, with accuracy that exceeds what human teams can deliver,” says Leigh Marie Braswell, Partner at Kleiner Perkins.
The bigger picture
The funding will be used to expand Poetic’s forward-deployed engineering team, enter new industries, and scale within its existing financial services customer base.
The company was profitable in 2025 on an eight-figure run rate with four employees. Headcount is expected to grow significantly as it moves into new sectors. Total funding to date, including the seed round, has not been disclosed.
The deeper question is whether Poetic’s approach holds as it moves beyond financial services into industries with different kinds of complexity — healthcare authorisations, government processing, supply chain decisions.
The company has chosen depth over breadth so far. Whether that model scales is the real test ahead.
