London-based startup Composo has raised $2 million in pre-seed funding to tackle a critical bottleneck in the adoption of enterprise AI: ensuring the accuracy and quality of AI-powered applications. The company's custom models offer a solution to the reliability issues plaguing large language models (LLMs), which are increasingly being used to power various applications.

Composo's approach is distinct from its competitors, such as Agenta, Freeplay, Humanloop, and LangSmith, in that it provides both a no-code option and an API. This dual approach widens the scope of its potential market, allowing domain experts and executives to evaluate AI apps for inconsistencies, quality, and accuracy without requiring developer expertise.

In practice, Composo combines a reward model trained on the output a person would prefer to see from an AI app with a defined set of criteria specific to that app. This system evaluates outputs from the app against those criteria, enabling clients to set custom guidelines for evaluation. For instance, a medical triage chatbot can have its client set custom guidelines to check for red flag symptoms, and Composo can score how consistently the app does it.

The company has recently launched a public API for Composo Align, a model for evaluating LLM applications on any criteria. This development has attracted notable clients such as Accenture, Palantir, and McKinsey, demonstrating the demand for Composo's solution.

According to Composo's co-founder and CEO, Sebastian Fox, the relatively low funding amount is due to the startup's approach not being particularly capital-intensive. Fox believes that the company's focus on solving the reliability issue will benefit from advances made by other companies building foundation models, such as OpenAI.

With the fresh funding, Composo plans to expand its engineering team, acquire more clients, and bolster its R&D efforts. The company's focus for the next year will be on scaling its technology across its client base.

Twin Path Ventures, a British AI pre-seed fund, led the seed round, with participation from JVH Ventures and EWOR. The investment is a testament to Composo's potential to address a critical bottleneck in the adoption of enterprise AI.

The reliability issue is a significant hurdle for the overall AI movement, particularly in the enterprise segment. As Fox noted, companies are now thinking beyond the hype and excitement of AI, asking whether it can truly change their business in its current form. Composo's solution aims to provide the necessary assurance and consistency for enterprises to confidently adopt AI-powered applications.

In terms of competitive advantage, Fox believes that the R&D required to develop Composo's solution is not trivial, and the company's first-mover advantage will be difficult to replicate. Additionally, Composo's approach is better suited to the rise of agentic AI, which is still in its early stages.

As the AI landscape continues to evolve, Composo's solution is poised to play a critical role in ensuring the accuracy and quality of AI-powered applications. With its unique approach and growing client base, the company is well-positioned to capitalize on the growing demand for reliable enterprise AI solutions.