Prentis, a new AI research lab co‑founded by serial entrepreneur Ritankar Das alongside Reid Hoffman and Mark Pincus, is negotiating to raise about $100 million at a roughly $1 billion valuation, according to two people familiar with the discussions.
What the company builds
Launched in April, Prentis trains models to understand how office workers carry out routine workflows across documents and systems. The aim is to build AI agents that can control computers and automate repetitive tasks — for example, handling insurance claims or processing customs duty refund exceptions without requiring humans to search for paperwork.
Those sources say Prentis has already signed contracts with an aggregate potential value of up to $50 million. Customers reportedly include a healthcare management service organization, a manufacturer, and goods and clothing producers. Investor materials reviewed by TechCrunch forecast an estimated $75 million annualized run rate by the third quarter of this year. The company’s pitch deck clarifies that these figures are based on an assumed contracted fee equal to 20% of realized savings (not recognized revenue) and are performance‑dependent and subject to final execution.
Claims about performance and cost
Prentis says its Hive‑32B model outperforms competitors — including OpenAI’s GPT‑5.4 and Anthropic’s Claude Opus 4.6 — on two computer‑use benchmarks: WindowsAgentArena, which measures end‑to‑end task completion in real Windows applications, and ScreenSpot‑v2, which tests a model’s ability to locate the correct on‑screen control. The company also claims its smaller model yields roughly ten times lower cost per task than frontier APIs, making it more economical to deploy across everyday workflows. TechCrunch has not independently verified these benchmark results.
Competitive landscape
Prentis is betting that automating everyday office tasks will soon eclipse coding as AI’s largest use case, but the space is crowded. Anthropic, OpenAI and Mira Murati’s Thinking Machines Lab are developing AI agents for computer use. Anthropic also acquired Seattle‑based computer‑use startup Vercept earlier this year, integrating its founders and winding down its product.
Founders and corporate background
Ritankar Das, Prentis’s CEO, is also the founder of Titan, a holding company that builds and operates AI businesses. Das, now 31, was UC Berkeley’s youngest University Medalist in more than a century, graduating at 18 with a double major in bioengineering and chemical biology, and later earned a master’s in biomedical engineering at Oxford. He founded Titan in 2014 after leaving a Cambridge AI PhD program where he was a Gates Cambridge Scholar.
Titan has launched and operated multiple companies, including Tala Health, an AI‑powered virtual care provider that closed a $100 million seed round last year, and Forta Health, an autism care startup that raised $55 million led by Insight Partners in 2024. Titan‑founded Dascena was acquired by CirrusDx in 2022.
For Prentis’s two other co‑founders the lab is more of a side project. Reid Hoffman, LinkedIn co‑founder and Greylock partner, said last month he would step down from Microsoft’s board after nearly a decade to return to “founder mode” on Manas AI, an AI drug‑discovery startup he also backs. Hoffman was an early OpenAI investor and co‑founded Inflection AI before much of that team moved into Microsoft in 2024. Mark Pincus, founder of Zynga, now runs the investment firm Reinvent Capital (with Hoffman as a senior adviser) and published a memoir, Life at the Speed of Play, last month.
Team and communications
Prentis’s website lists more than 25 employees, including researchers who previously worked at OpenAI, Google DeepMind, Meta, Tencent and Alibaba. The company did not respond to TechCrunch’s request for comment.
Why this matters
Prentis’s approach — deploying smaller, lower‑cost models targeted at specific, frequent enterprise tasks and charging based on a share of realized savings — highlights a possible commercial path for AI beyond large generalist models. If contracts hold and the promised savings materialize, the group’s work could signal a broadening of AI adoption into routine office automation. At the same time, several large players are pursuing similar goals, making the field competitive.
