Merlin Releases DiscoveryPartner: the FirstGenAI-Powered Discovery and Investigations Platform

[EDRM Editor’s Note: EDRM is happy to amplify our Trusted Partners news and events.] The opinions and positions are those of John Tredennick.]

• November 14, 2023
• Download the PDF

DiscoveryPartner™ offers a revolutionary way to find, review and analyze documents for investigations, ediscovery and other large data sets– achieving in minutes what typically takes hours or days, allowing clients to save on hosting costs for clients with On/Off Cloud Utility pricing™.

Denver CO, Nov. 13, 2023—Merlin Search Technologies, a pioneering AI and cloud technology company, announced the release of DiscoveryPartner, a full-featured Generative AI platform designed to reengineer search, investigation and discovery workflow. By harnessing large language models (LLMs) like GPT and Claude for critical discovery tasks, DiscoveryPartner slashes the time and costs of traditional discovery workflow, allowing clients to find, analyze and review large documents sets faster and at substantially lower costs than was ever possible using review teams alone.

DiscoveryPartner is delivered securely via the cloud, using unique single-tenant cloud architecture for greater security and flexibility, coupled with groundbreaking On/Off Cloud Utility Pricing. Clients can reduce hosting costs by up to 70% by switching sites off when not in use, while reducing their carbon footprint.

A Singular Vision

“Merlin started out with a singular vision—to revolutionize investigation and discovery workflows through AI and advanced cloud technologies,” explained Merlin’s founder and CEO John Tredennick. “Our objective was to make search, analysis and review of large document sets faster and more effective, while reducing the costs associated with traditional, labor-intensive workflow. After three years of development, a million lines of code and 100 platform releases, we’re doing just that. DiscoveryPartner is a game changer for investigations and discovery, saving clients and their law firms millions of dollars.” 

Tredennick began his career as a trial lawyer and partner at a national law firm. He and his team began developing litigation technology in the mid-90s, before founding Catalyst Repository Systems, which became a leading international ediscovery provider. In 2019, after selling Catalyst to a large public company, Tredennick founded Merlin Search Technologies to take document investigation and discovery to the next level. He brought his India-based development team (with whom he had worked for almost two decades) and a senior data scientist who had worked with him at Catalyst.

Powered by Sherlock AI 

DiscoveryPartner is powered by Sherlock AI, a set of generative AI and machine learning algorithms that automate critical discovery tasks, allowing them to be completed in minutes, rather than hours or days.

Clients can use Sherlock AI to:

• Find documents without complex keyword searches. Sherlock AI keys off user judgments, analyzing and ranking a subset or the entire data set in milliseconds.
• Analyze individual and large groups of documents to quickly identify key witnesses, issues and information and synthesize content across documents.
• Review documents with GenAI, saving on professional costs and time, with the algorithm tagging documents for responsiveness, providing the reasons for each classification.

With a simple prompt, Sherlock AI can identify key witnesses, summarize document content, create timelines, suggest deposition candidates, provide questions and create full-blown investigation reports. All information can be downloaded to create reports.

Sherlock AI also can summarize and analyze hearing and deposition transcripts, allowing users to ask questions, analyze relationships, create timelines and even look for contradictions in testimony. Merlin is in the process of adapting it to handle large volumes of SMS and chat messaging, searching, analyzing and reporting on multi-party conversations in seconds. 

Watch Sherlock AI find, analyze and review documents, providing summaries and answering questions, in this short video.

A Multi-LLM Approach

Merlin chose a multi-LLM approach for DiscoveryPartner, allowing users to choose from a number of GenAI engines for different analytical tasks and pricing structures. 

“We chose to build a multi-LLM framework because we have found that lower-priced models can handle certain tasks as well as the higher-priced versions, while completing them much more quickly,” explained Dr. William Webber, Merlin Chief Data Scientist. “With a multi-LLM framework, our clients can mix and match LLMs to get the most cost-effective and timely results based on the task–using faster and cheaper models for routine work while reserving the more sophisticated ones for synthesis, analysis and reporting,” he added.

“As new and even more powerful LLMs hit the market, we can quickly integrate them into the platform,” added Webber. “Our adaptable framework ensures that we remain at the forefront of technology, offering clients an agile, powerful and economical tool that elevates their practice in an increasingly competitive and technologically-driven environment.”

Reengineering Cloud Architecture for Greater Security

DiscoveryPartner is deployed via a unique single-tenant cloud infrastructure. Each site resides in its own virtual private cloud (VPC) and runs on separate hardware and software. Single-tenant architecture provides greater security and flexibility, eliminating data commingling and allowing for localized hosting anywhere in the world to meet global data privacy regulations. 

“Single-tenant architecture gives clients more control over their sites, allowing us to configure sites for different data volumes and to configure backup and disaster recovery capabilities to meet individual client needs. We can deploy a new site quickly, positioning it virtually anywhere, including in the client’s own AWS cloud environment,” stated Kathy Hogy, Merlin Chief Client Officer. 

Merlin’s approach is different from other discovery vendors, who typically host client data in a multi-tenant environment. “We were concerned about commingling client data through shared computing resources,” Tredennick explained. “If one site were compromised through ransomware, for example, we didn’t want put other client data at risk.”

On/Off Cloud Utility Pricing to Save on Hosting

Along with increased security, Merlin’s single-tenant architecture set the stage for a new way to charge for hosting, billing by the minute rather than the month. Clients pay one hourly rate when they are using the site and a 70% lower rate when they turn it off. Contrast this with traditional pricing where clients pay by the month whether they use the site or not.

Clients save on hosting because discovery sites are rarely in use for more than a fraction of the month. “There are 730 hours in a month,” explained Tredennick. “To reach 25% active use, the team would have to be online for eight hours a day, five days a week for the entire month. That might happen during an active review period but not for the many other months a case is typically online.”

Cloud Utility Pricing allows clients to control hosting charges. Administrators can turn off the site through DiscoveryPartner’s administration module. They can also schedule operating hours or use Merlin’s unique Cloud Energy Saver to automatically turn off the site when not in use. These features alone can cut hosting costs by half, even on a site undergoing active review. Clients can also shut down sites for the entire month when the matter is dormant. Any user with the site passcode can turn it back on in minutes.

The First Green Computing Initiative 

Merlin’s Cloud Utility Pricing model also constitutes the first green computing initiative for investigations and discovery. By turning off sites, clients save on energy use as well as hosting costs. In minutes, all of the computing resources, from front-end web servers to the database, full-text engine and AI servers, are shut down, reducing energy consumption while the sites are offline.

DiscoveryPartner’s Cloud Energy Saver monitors each site, watching for periods of  inactivity. When the system determines that a site has not been active for a set amount of time, it turns the site off, saving energy and reducing hosting costs. “This won’t keep users from accessing the site when they need it,” Tredennick explained. “We are just reducing energy consumption and saving on computing costs when they aren’t using the site.”

“We are proud to play a part in reducing our carbon footprint and combating climate change,” Tredennick stated. 

Continued Innovation

While Build 100 represents an important milestone for Merlin, the team will continue to innovate, exploring the boundaries of generative AI and any new technology that follows. “Our experiments indicate that, with an appropriate topic description, advanced LLMs are at least as reliable as humans at performing first-level responsiveness review. We’re in the midst of developing LLM-driven workflows for building, refining, and deploying such topic descriptions,” stated Webber. “We’ll continue to test new ways to use GPT and other LLMs to help our clients get through their work faster and more cost-effectively.”

“Our work is far from done,” concluded Tredennick. “We’ll continue monitoring developments in GenAI, while integrating new cloud technologies that make our clients more efficient and effective, while making sure DiscoveryPartner continues to deliver on its core promises of speed, security and savings.” 

Join Us in New York the Week of November 27th

Tredennick will be in New York City the week of November 27th to introduce DiscoveryPartner to law firms and corporations. 

On November 29th at 12:00pm, Tredennick and  Dr. Webber will present an in-person masterclass, Using Generative AI to Transform Investigations and Discovery Workflow. The event is quickly filling but there is still time to register here. 

Tredennick is also scheduling meetings with a limited number of law firms and corporate legal departments between Tuesday and Thursday that week. To schedule a meeting, send him an email. 

To Learn More About DiscoveryPartner

• Visit Merlin.Tech or tune in to Merlin’s Generative AI web page to learn more about the benefits of harnessing LLMs for investigations and discovery.  
• Reach out for a personal demo with an email to Sherlock@Merlin.Tech or contact us through our website.

Read the original release here.

About Merlin Search Technologies

Merlin is a pioneering cloud technology company leveraging generative AI and cloud technologies to re-engineer legal investigation and discovery workflows. Our next-generation platform integrates GenAI and machine learning to make the process faster, easier and less expensive. We’ve also introduced Cloud Utility Pricing, an innovative software hosting model that charges by the hour instead of by the month, saving clients substantial savings on discovery costs when they turn off their sites. With over twenty years of experience, our team has built and hosted discovery platforms for many of the largest corporations and law firms in the world. Learn more at Merlin.Tech.

Media Contacts