AWS makes it easy for us, as a small company, to focus on maturing our application rather than spending time on infrastructure as we continue to grow our business.
Eric Alexander Chief Technology Officer, LinkSquares
  • Challenge

    LinkSquares seeks to disrupt the legal and finance industries by providing an automated, software-based solution to streamline post-signature contract analysis. The company grew quickly and needed to optimize its existing solution in order to conduct contract analysis efficiently and accurately at scale.

  • Solution

    LinkSquares engaged with SFL Scientific to build a custom machine learning solution on AWS.

  • Benefits

    By running a machine learning solution on AWS, LinkSquares can focus its time and resources on developing new solutions and solving even greater business problems for end users.

LinkSquares provides high-growth companies with a suite of tools to complete fast and systematic legal reviews of executed business agreements. Through contract analysis and reporting, LinkSquares’ tools help relieve businesses from the burden of manually and inefficiently reviewing contracts. Whether a client is preparing for an audit, reviewing agreements for a merger, or responding to a crisis, LinkSquares’ cloud-based solutions help businesses make accurate decisions at an accelerated rate, providing company stakeholders with increased time and assurance.

Companies experiencing rapid growth often lack the bandwidth to track each line of every contract, service agreement, or legal document before it’s executed. Even in the most carefully reviewed agreements, some information is forgotten as soon as the contract is signed. Once the business has matured and due diligence projects arise (for example, when a law changes or an acquisition takes place), companies must conduct detailed reviews of all signed contracts and identify specific terms within them.

LinkSquares’ founders experienced the painful reality of reviewing existing legal contracts firsthand when their previous employer underwent an acquisition. “We manually searched through all of our existing contracts to identify privacy language and other information crucial to legally moving clients to a new infrastructure provider for our software service,” says Vishal Sunak, co-founder and chief executive officer at LinkSquares. “We found that the language deviated from contract to contract, making the review process of thousands of agreements both timely and frustrating.” The team identified existing software solutions helping companies efficiently address the pre-signature workflow: contract creation, terms negotiation, and internal workflow. However, the industry lacked a software solution to help companies mine for information in existing contracts. LinkSquares saw this gap as an opportunity to develop software to help customers with post-signature contract analysis.

“We sought to disrupt an older industry historically focused on pre-signature work. We’re focused on post-signature analysis and we don’t deal with anything pre-signature,” says Eric Alexander, chief technology officer at LinkSquares. “We chose to build a software as a service offering on AWS so that we can get companies migrated from their existing storage solutions and up and running using our software quickly, enabling them to understand what they agreed to in their contracts.”

Even with the ability to quickly mine for contract data in a cloud-based environment, the LinkSquares team still needed a scalable solution for identifying and classifying legal language. Initially, they built a searchable contract database, but as the company grew, so did its need for an automated solution for extracting key contract metadata.

The LinkSquares team turned to the experts at SFL Scientific, an AWS Partner Network (APN) Consulting Partner and AWS Machine Learning Competency Partner. The team at SFL Scientific builds deep relationships with each client to understand the client’s challenges and its short- and long-term vision for using artificial intelligence and machine learning technologies. SFL Scientific excels in its ability to help clients execute data-driven strategies. The company worked closely with LinkSquares to understand their current pain points and future goals.

“LinkSquares built its initial prototype using SQL running on AWS, but it wasn’t using any machine learning technology,” says Michael Luk, chief technology officer at SFL Scientific. “We learned about the team’s vision to process terms and language automatically and understood the business pain points. We proposed building a custom machine learning solution to help the team scale and improve accuracy.”

One month after its initial engagement, data scientists at SFL conducted a proof of concept showing the LinkSquares team how they could deploy a scalable, automated analysis solution using machine learning on AWS.

SFL Scientific used natural language processing (NLP), an artificial intelligence (AI) method helping computers understand and interpret human language, to build its machine learning algorithm. Implementing the algorithm enabled LinkSquares’ software to extract key terms from a document and tokenize these terms into pre-defined categories. Upon deployment on AWS, the algorithm ran the code on demand. Whenever a document was uploaded, the machine learning code automatically launched.

The text extraction algorithm process consisted of three main steps: feature engineering, model stacking ensemble, and post-processing. First, the algorithm parses raw text and stores it individually. Next, tokenized texts create hundreds of unique features based on rule-based features, token-based features, and sequence-level classes as features. After the feature engineering, a model stacking ensemble technique predicts the class of a token. The modeling, which was trained against the human-tagged data, assigns a probability to each class prediction, making it possible to determine a probability threshold (or level of guaranteed accuracy). Finally, once the classes for each token are predicted and cleaned, continuous tokens are strung together, making it easier to digest the data. Click here to learn more about the SFL Scientific’s algorithm.

The NLP algorithm developed by SFL completely revolutionized the post-signature contract review process for LinkSquares. The machine learning code enables the LinkSquares software platform to automatically run code on thousands of documents in seconds. Every result showed an exponential improvement in time spent reviewing each document, and eventually, improvement in tagging accuracy compared to the human auditors.

Using AWS to power its document storage, automated search, and machine learning capabilities lets LinkSquares focus its resources on optimizing products rather than maintaining infrastructure. “AWS makes it easy for us, as a small company, to focus on maturing our application rather than spending time on infrastructure as we continue to grow our business,” says Alexander.

Having identified SFL Scientific to help it take advantage of machine learning technology further emboldens LinkSquares as the team develops cutting-edge solutions on AWS. “It’s been fantastic engaging with SFL Scientific as its team are experts in the AI space,” says Alexander. “They understand the business challenges we’re trying to solve and they’ve given us the guidance we need to use new technology to tackle these challenges. I think of them like they’re a part of our team. They’re a valued partner.”

LinkSquares’ AI-powered solution on AWS is a fundamentally new approach to the streamlining of post-signature contract analysis. The team plans to explore additional technologies on AWS that they can use to drive further innovation in their industry. “We’re excited about the future of our offering and how we can help legal and finance teams eliminate manual reviews of files,” says Sunak. “We’re excited to build out more AI and take advantage of new AWS services to continue exploring what’s possible.”

SFL Scientific is a data science consulting and professional services firm. The company uses specific domain knowledge to solve complex, novel, and R&D problems, specializing in helping develop a fully integrated approach to leveraging data-driven systems and improve decision-making with AI. SFL Scientific creates end-to-end solutions and offers its customers customization through data strategy and technological development.

Connect with SFL Scientific