Introducing @PoeticHQ: a new AI system that executes complex multi-hour tasks with 99% accuracy and 10x fewer tokens than agents. We raised $50M at $500M from Kleiner Perkins, Founders Fund, First Harmonic, and Genius Ventures to build AI that does complex work inside Fortune 500 companies without hallucination. While code is too brittle, agents are too unpredictable. The work that runs the global economy - anti-money laundering, fraud investigations, underwriting - needs extreme accuracy. So we built a new kind of software that pairs the flexibility of AI with the predictability of code. When the world stays the same, Poetic runs fixed code: fast, cheap, identical every time. When the world changes, Poetic uses AI to regenerate its approach and find its way back to the objective. In one year, we went from zero to an eight-figure run rate as a team of four. Since then, we’ve scaled the team and executed the highest-stakes processes at AIG, SoFi, and Chime. At SoFi, a large US bank, Poetic reached 99% quality on fraud investigations in five weeks.

Putting members first isn't just a core value—it's how we build. Together with @PoeticHQ, we're using AI to streamline fraud operations, helping drive faster resolutions and greater consistency while maintaining high standards for our members. Congrats on the milestone. 🚀

Stepping outside my Stanford/Google/Waymo research bubble and sitting with real American companies made me realize why the bulk of the economy was not automatable by software or AI. We needed something new that is neither strictly code or AI, something that can flex but deliver reliable and affordable results to the biggest companies in the world. @tbpn @PoeticHQ

Return on Tokens = (Value of Output - Cost of Tokens) / Cost of Tokens x 100 Thinking is expensive but happens rarely. Doing is cheap and happens forever. Agents should do the thinking, code should do the doing. From the @markiewagner @packyM essay. notboring.co/p/return-on-tok…

@MarkieWagner explaining to @TBPN how Poetic is able to run multi-hour complex processes for F500s with 99% accuracy in production, all with minimal token spend.

Introducing @PoeticHQ: a new AI system that executes complex multi-hour tasks with 99% accuracy and 10x fewer tokens than agents. We raised $50M at $500M from Kleiner Perkins, Founders Fund, First Harmonic, and Genius Ventures to build AI that does complex work inside Fortune 500 companies without hallucination. While code is too brittle, agents are too unpredictable. The work that runs the global economy - anti-money laundering, fraud investigations, underwriting - needs extreme accuracy. So we built a new kind of software that pairs the flexibility of AI with the predictability of code. When the world stays the same, Poetic runs fixed code: fast, cheap, identical every time. When the world changes, Poetic uses AI to regenerate its approach and find its way back to the objective. In one year, we went from zero to an eight-figure run rate as a team of four. Since then, we’ve scaled the team and executed the highest-stakes processes at AIG, SoFi, and Chime. At SoFi, a large US bank, Poetic reached 99% quality on fraud investigations in five weeks.

The first time I heard @markiewagner discuss this idea we were sitting around a campfire at something like 2am just months before the public launch of ChatGPT. Watching her pursue her Good Quest and level-up as a hero Paladin on this mission has been incredibly fun. She's become not only one of my favorite founders, but also one of my favorite people and friend. Congrats on the launch today, Markie! 🚀

In November 2022, @markiewagner wrote Choose Good Quests, then she went dark to work on her own. Today, she's launching Poetic, a new class of software that's adaptive like AI, reliable like code. This is her first public essay since CGQ, on why & how. notboring.co/p/return-on-tok…

Agents should do the thinking, code should do the doing.

At a moment when companies are starting to question whether AI token spend is worth it, @PoeticHQ is already offering the solution. Near-tokenless agents, 99% accurate on the workflows that actually run large companies - deployed at the world's leading banks and insurance agencies. Customers include SoFi, Chime, and AIG. Thrilled to partner with @markiewagner and this small, but mighty team.

Markie and team are legends. Thrilled to see @PoeticHQ out in the world.

Markie is one of the most prescient people I know on AI, and I've been lucky to have a front row seat since the beginning. @PoeticHQ hit an 8-figure run rate and became profitable in 2025 with just 4 people. At a moment when token ROI is under scrutiny everywhere, @markiewagner and the Poetic team have already built a viable solution. KP blog with the story behind the investment - kleinerperkins.com/perspecti…

If you want to teach AI to do real work inside the world’s top companies, join us. poetic.com/careers

Thank you Bloomberg, for covering our story.

We should choose good quests.  Ones that lead to prosperity for the American worker and a stronger nation.    AI shouldn’t be about frivolous pursuits that further fracture our attention and connection to humanity.  We are a great nation with serious challenges, amongst them the CCP waging a war to supplant and subjugate us and the legitimacy/capacity of our institutions, public and private.  To win we must reindustrialize, produce more energy, empower the American worker – and use our asymmetric advantage in AI to do it.  Frivolous applications of AI do nothing to address the greatest challenges of our time and adds more weight to the growing anti-AI movement which sees America’s AI companies as disinterested in their fate, exploitative, and craven.  What are the outcomes? Whose life is being improved?  What are you delivering to the American people and the nation?  These are the essential questions that all of us working on AI must answer.

We asked @traestephens about his "Good Quests" essay with @markiewagner. "Lots of super talented former entrepreneurs took the route of Big Head in Silicon Valley (HBO show) to rest and vest." "That feels like a moral failure of tech. Founders operating on God Mode should focus on the hardest and most important quests, like rebuilding semiconductor manufacturing in the US."

A little over two years ago, @markiewagner and I published an essay in @PirateWires entitled "Choose Good Quests" that blew up way more than I was expecting. I was also surprised that more than a few readers were able to read between the lines to see the theological underpinnings of the central argument. I'll be giving a talk hosted by ACTS 17 @garrytan on the evening of March 6th in San Francisco in which I'll draw out the explicit theological message and it's implications on how we think about our quests in tech. Invite link in thread below. Should be fun!

Choose Good Quests @markiewagner

A manifesto for founders in tech: Choose Good Quests, by @traestephens and @markiewagner "There are certainly bad quests that are hard, and (a few) good quests that are easy." Read it here 👇🏻