HerculesAI: Pioneering Large Language Models Long Before They Became Trendy

HerculesAI, previously known as Zero Systems, has been at the forefront of automating professional services since 2017, initially focusing on the legal sector. The company has been developing large language models (LLMs) for several years, long before they became a hot topic in the tech world. This expertise positioned HerculesAI perfectly to leverage the surge in interest around LLMs following the release of ChatGPT in late 2022. As a result, the company recently secured a $26 million Series B investment to further capitalize on this momentum.

Alex Babin, CEO and co-founder, reveals that they began working on smaller models with parameter counts ranging from 500 million to 2 billion around 2020, deploying these models on edge devices to ensure compliance. However, prior to ChatGPT's rise, there was little interest from decision-makers regarding this aspect of their technology. “About eight or nine months before ChatGPT launched, I was explaining to CIOs what an LLM could do — and no one was interested,” Babin recounted. That landscape changed dramatically by November, sparking widespread curiosity about LLMs and driving substantial growth for the company over the past year.

Currently, HerculesAI's models support three essential functions: intelligent data extraction, data transformation, and data verification. Data extraction involves retrieving information from documents, while transformation automatically organizes this data into structured rules. Most crucially, verification ensures that any discrepancies between the extracted data and the original sources—such as regulations, policies, or contracts—are flagged for users. “It’s really the holy grail to compare the extracted information to the source of truth,” Babin emphasized, highlighting its significance in maintaining accuracy and compliance.

These capabilities have also allowed HerculesAI to create a multi-agent system to automate these processes. “Our multi-agent systems can be applied to high-value, continuous workflows that require automated decision-making,” he explained. This functionality is particularly critical for their core clients in regulated industries, which now include not just legal services, but also the insurance and financial sectors.

The company’s AI-focused strategy is proving effective, with HerculesAI experiencing a remarkable 4x growth over the past year. Impressively, 30% of the top 100 law firms in the U.S. now count themselves among their clientele, alongside major Fortune 500 companies such as Mercer, Standard & Poor’s, and State Farm.

Although HerculesAI has around 75 employees, Babin shares that the recent funding will be used to enhance internal processes rather than significantly expand the workforce. “I don’t see the need to hire more staff. We will focus on refining our internal processes and automation. We have to eat our own dog food and use our own products to scale effectively,” he stated.

This latest funding round was led by Streamlined Ventures, with contributions from Proof VC, Thomson Reuters Ventures, Alumni Ventures, and various industry angels.