Hypercell for GenAI transforms complex documents into LLM and RAG-ready data, accurately, automatically, and continuously
Solution provides business user ready interface to label and annotate high-quality enterprise data, and interact with RAG and GenAI to drive automation and decisioning
Hyperscience is partnering with organizations including Google Cloud and Hewlett Packard Enterprise to provide an end-to-end offering for applying foundation models based on enterprise data embedded at the core of businesses
New York, NY – June 5, 2024 – Hyperscience, a market leader in hyperautomation and a provider of enterprise AI infrastructure software, today announced a new solution that ushers the back office into the GenAI age, by fine-tuning LLMs with ground truth documents embedded at the core of the enterprise. Hypercell for GenAI automatically annotates, labels, and structures data from documents for fine-tuning LLMs and GenAI experiences, allowing organizations to rapidly and continuously develop highly accurate, relevant and valuable enterprise models. Through a trusted and proven interface, business people can use Hypercell for GenAI to accelerate mission critical workflows, grounded in secure, proprietary data, and tuned to the business.
Hyperscience is working with Google Cloud, Hewlett Packard Enterprise (HPE), and other partners on this solution, to give customers flexibility to operate in the infrastructure and AI development platform of their choice, to enable use cases such as prompt engineering, RAG, grounding, and vector search on the customers’ proprietary enterprise data.
“The success or failure of any AI initiative starts with the data that feeds the models,” said Andrew Joiner, Chief Executive Officer, Hyperscience. “Too often, models are built on faulty and incomplete data, and inefficient manual methods and legacy technologies struggle to keep pace with the dynamic flow of documents that course through organizations every day. Today, Hyperscience provides a breakthrough to this challenge by allowing organizations to establish an accurate data estate that trains LLMs to speak the language of their business, and empowers users with relevant, in-context GenAI experiences that align with their business processes and use cases.”
“As organizations rush to harness the power of large language models, their data scientists need abundant, top-notch data to train these models effectively,” said Alan Pelz-Sharpe, Founder and Principal Analyst of Deep Analysis. “Much of this valuable data is locked away in unstructured documents, making it unusable in its current format. Hyperscience’s latest solution aims to address this challenge head-on by preparing unstructured data, ensuring it’s accurately labeled and annotated for language model training. With its powerful engine and scalable platform, Hyperscience is well positioned to tackle this crucial task, paving the way for successful in-house LLM deployments.”
Extending Hyperautomation to GenAI
Over the past decade, organizations have pursued digital transformation initiatives in order to automate, modernize, and compete in a dynamic and fast-moving marketplace. Traditional Intelligent Document Processing (IDP), Robotic Process Automation (RPA), and Optical Character Recognition (OCR) technologies have failed to deliver on the promise of digital transformation, since these offerings are rigid and rules-based, and struggle to adapt to new, varied, and complex documents inside an organization. These solutions have delivered sub-par performance in accuracy, and require significant, expensive manual effort from business process outsourcers (BPOs) to automate processes and workflows inside organizations.
Hyperscience has disrupted this paradigm with a novel approach built on AI at its core. Based on a proprietary, machine learning model-based architecture that reads and understands content fluently, Hyperscience delivers industry-leading accuracy rates of 99.5% and automation rates of 98%. Designed with simplicity, business users can train and manage models based on their domain expertise, and the Hyperscience platform’s blocks and flows enable process orchestration and integration with downstream enterprise applications. This has unleashed new levels of productivity for organizations, allowing them to connect end-to-end processes and decisioning from the back office to the front office, and positions Hyperscience as a market-leader in the fast-growing hyperautomation market.
Today, Hyperscience is extending hyperautomation to GenAI in the enterprise. Up until now, LLMs have primarily trained on publicly available data, but have struggled to access and deliver accurate results on data and documents in the back office, like forms, invoices, insurance claims, bills of lading, and more. Hypercell for GenAI leverages the same core technology for hyperautomation, to rapidly transform complex documents into LLM and RAG-ready data, accurately, automatically, and continuously.
Teaching LLMs the language of your business
Hypercell for GenAI establishes a comprehensive data estate to power relevant, in-context GenAI experiences. The solution provides a simple user interface that delivers trusted, accurate results as part of a business user’s workflow.
For example, an insurance claims adjuster could use the Hypercell for GenAI to ask questions in a natural language prompt on the status of a claim. The solution can convert complex documentation such as forms, medical reports, receipts, and doctor’s notes into RAG-ready data for summarization, and provide a recommendation to users on whether to approve or reject the claim based on this ground truth data.
Flexibility and choice in infrastructure environment and AI development platform
Hypercell for GenAI can run on-premises, in a hybrid cloud, in a public cloud, in a SaaS environment, and even highly secure air-gapped environments. Built with the highest security, governance, compliance, and traceability standards, the offering is well suited for use cases in regulated industries and the public sector. Hypercell for GenAI provides a cutting edge architecture that enables organizations to deliver enterprise AI cost effectively, on top of CPUs or GPUs. The solution supports a wide range of LLMs, including Mistral Large and Mistral 8X22B, Llama3 (including all three versions), and GPT 3.5 and 3.0.
Hyperscience is collaborating with Google Cloud and Hewlett Packard Enterprise, leveraging AI software that accelerates AI model development by transforming complex documents embedded at the core of the enterprise into an accurate, AI-ready data estate for applications, including generative AI.
Global Campaign – Hypercell for GenAI
Hyperscience launched Hypercell for GenAI today as part of a global campaign, including billboards in London, Salt Lake City, Sofia, Bulgaria, and Times Square in New York City. Hyperscience also today announced strong momentum, including record Q1 2024 results, record customer retention rates, and strategic customer wins.
Availability
Hypercell for GenAI will be available on Google Cloud Marketplace or directly from Hyperscience. Hypercell for GenAI leverages the core capabilities of the Hypercell, an all-in-one AI infrastructure software platform that accelerates enterprise AI initiatives at scale.
About Hyperscience
Hyperscience is a market leader in hyperautomation and a provider of enterprise AI infrastructure software. The Hyperscience Hypercell platform unlocks the value of an organization’s back office data through the automation of end-to-end processes, and transforms complex documents into LLM and RAG-ready data to power new enterprise GenAI experiences. This enables organizations to transform manual, siloed processes into a strategic advantage, resulting in a faster path to decisions, actions, and revenue; positive and engaging customer, public, and patient experiences; and dramatic increases in productivity.
Leading organizations across the globe rely on Hyperscience to drive their hyperautomation initiatives, including American Express, Charles Schwab, Fidelity, HM Revenue and Customs, Mars, Stryker, The United States Social Security Administration, and The United States Veterans Affairs. The company is funded by top tier investors including Bessemer Venture Partners, Battery, FirstMark, Stripes, and Tiger Global.
Media Contact
Jyotsna Grover
415-917-7411
[email protected]
