Maxim AI Raises $3 Million To Put GenAI To The Test

A new technology requires a new approach to testing, argue the founders of Californian start-up Maxim AI, which is today announcing it has raised $3 million of new capital. The business, launched in 2023, has developed a new platform aimed at developers of applications that harness generative artificial intelligence (GenAI) – particularly those in medium-sized and large enterprises.

“We want to make it possible for developers to test and ship their products much faster,” says Vaibhavi Gangwar, a former Google staffer, who co-founded the business with Akshay Deo, who previously worked at the app development platform Postman. “In a post ChatGPT world, the needs and challenges related to testing are very different.”

Traditional software products rely on a well-established testing process, Gangwar points out. The developer has a clear view of how the application operates and can test its ability to perform specific tasks without error or breakdown.

By contrast, GenAI products are non-deterministic – that is, they deliver unpredictable outputs according to a range of factors that developers struggle to control and sometimes to identify. Testing such products can therefore be a time-consuming process that requires continuous human intervention.

The stakes may also be higher with many GenAI products, given concern about hallucinations, inaccuracies and ethics. Organisations that do not commit the resources required for rigorous testing may compromise the user experience; they also face significant reputational risk.

Maxim’s platform therefore aims to provide GenAI developers with a piece of infrastructure through which they can standardise their testing processes. The platform is embedded in the development process, enabling developers to test their products from the beginning of engineering onwards to post-release fine-tuning. It also enables human evaluation, which is often critical at certain stages of GenAI app development.

Gangwar believes the platform has the potential to rapidly accelerate app deployment. “We’re able to boost developer productivity to such an extent that their apps can ship up to five times’ more quickly than they do currently,” she claims.

Deo argues that the benefits go well beyond faster shipping. “For the benefits to be realised meaningfully up the value chain, we need the products built on top of the generative AI stack to be trustworthy and reliable,” he says. “That makes evaluation a very critical piece of the AI infrastructure.”

Still, it’s early days for Maxim. While the business has attracted a number of paying customers – particularly from the software-as-a-service sector – it remains in private beta mode for now. It also faces competition from others in the testing space, though the founders argue that they offer a more integrated solution.

Today’s fundraising should allow the company to put such claims to the test in a more meaningful way. The proceeds of the raise will be deployed to drive product development, recruitment and third-party partnerships; Maxim will also now move from private beta to general availability.

The $3 million raise is led by Elevation Capital, with participation from a number of business angels. “Maxim AI is addressing a crucial need in the evolving AI landscape,” says Akarsh Shrivastava, a principal at Elevation Capital. “The focus on creating a standardised, scalable evaluation infrastructure addresses a deep pain point for AI developers; as the complexity and diversity of AI use cases continues to increase, Maxim’s platform will play a critical role in ensuring trustworthy and reliable AI systems.”

Indeed, Maxim’s founders believe that a nascent technology now requires a more robust approach to testing. “All of us building products in this industry are working in an immature space, even if it is evolving at tremendous speed,” adds Gangwar. “There is a tremendous value creation opportunity in GenAI, but we need to think hard about how we can be more productive and collaborative.”