Anthropic Welcomes Humanloop Team as Competition for Enterprise AI Talent Heats Up

Anthropic has acquired the co-founders and most of the team from Humanloop, a platform focused on prompt management, LLM evaluation, and observability, as part of its strategy to enhance its enterprise services.

Although the details of the transaction remain confidential, it appears to follow the growing trend of acqui-hiring in the tech industry, driven by the increasing competition for AI talent. Humanloop’s three co-founders—CEO Raza Habib, CTO Peter Hayes, and CPO Jordan Burgess—along with around a dozen engineers and researchers, have joined Anthropic.

Anthropic is rapidly expanding within the enterprise sector, demonstrating strength in agent-based and coding functionalities. A representative from Anthropic confirmed that the company did not acquire the assets or intellectual property of Humanloop; however, in an environment where expertise is often as valuable as ownership, this detail may be less significant. The experience brought by Humanloop’s team will be vital in developing tools that allow enterprises to implement safe, reliable AI solutions at scale.

Brad Abrams, API product lead at Anthropic, remarked: “Their established expertise in AI tooling and evaluation will be essential as we enhance our efforts in AI safety and the development of functional AI systems.”

In a competitive environment where merely having a superior model is not enough, improving its tooling ecosystem may give Anthropic a solid advantage over OpenAI and Google DeepMind, both in performance and readiness for enterprise applications.

Humanloop, founded in 2020 as a spinout from University College London, participated in Y Combinator and the Fuse Incubator, and has since raised $7.91 million in seed funding in two rounds led by YC and Index Ventures, according to PitchBook. The company has built a strong reputation for helping enterprise clients—including Duolingo, Gusto, and Vanta—in developing, evaluating, and refining robust AI applications.

Just last month, Humanloop informed its customers of plans to shut down in anticipation of the acquisition.

The timing of this acqui-hire aligns with Anthropic’s efforts to provide features like extended context windows for enterprise clients, which enhances the capability and applicability of its models.

Earlier this week, Anthropic secured a contract with the central purchasing arm of the U.S. government to deliver its AI services to various agencies across the executive, judiciary, and legislative branches for just $1 per agency in the first year— a clear tactic to compete with OpenAI’s similarly priced offerings. Both government and enterprise clients need the evaluation, monitoring, and compliance capabilities that Humanloop specialized in.

This acquisition fits well with Anthropic’s identity as a “safety-first” AI company. Humanloop’s evaluation workflows support this mission by providing ongoing performance assessments, safety measures, and bias mitigation.

“From the beginning, we have concentrated on creating tools that empower developers to build AI applications safely and efficiently,” remarked Raza Habib, former CEO of Humanloop. “Anthropic’s commitment to AI safety research and responsible AI development aligns perfectly with our vision.”

We’re always looking for ways to improve, and your feedback on TechCrunch and our events can assist us! Complete this survey to let us know how we’re doing and have a chance to win a prize!