Flipping the Script: Koobrik’s Journey to Unearth Cinematic Treasures with GenAI and AWS



  • 2 February 2024
Share this post
AWS Funding Secured by Cloud303
  • Well-Architected

About the Customer

Koobrik is an innovator in creative data management, specializing in transforming institutional memories into instantly searchable resources. They streamline collaboration within entertainment organizations and design custom project tracking and data solutions.

Executive Summary

In the vibrant realm of Hollywood, where narratives become the lifeblood of culture and the seeds of dreams, Koobrik arose as a harbinger of a new epoch. This case study chronicles the odyssey of Koobrik as it harnessed the formidable prowess of Amazon Web Services to redefine the screenplay development process, heralding an avant-garde era in the art of cinematic storytelling.

Koobrik's quest was born out of a critical industry impasse: the vast majority of screenplays, rich with potential, lay dormant, their stories untold, due to the sheer magnitude of data. In response, Koobrik embraced AWS's suite of technologies to construct an innovative ecosystem that could intelligently and meticulously parse, analyze, and archive the essence of film scripts. By ingeniously integrating GenAI services of Amazon Bedrock for text generation, Amazon Textract for text and data extraction, Amazon Comprehend for content analysis, and Amazon SageMaker for machine learning, Koobrik established a groundbreaking platform. This platform not only democratizes the evaluation process of screenplays but also augments the creative capacity of the industry, enabling producers, writers, and executives to unlock the full potential of their content.

The Challenge

Koobrik's mission was to tackle a formidable antagonist: the overwhelming and ever-growing volume of creative screenplays. This deluge of scripts, an outpouring of creativity and aspiration, led to a startling reality - an alarming 80% of these screenplays languished, unread and unrecognized. The potential cinematic gems - each a universe of ideas and emotion - remained undiscovered, obscured by the overwhelming volume of data.

The industry, with its legacy systems, was straining under this load. As Orlando Wood, a film major from Columbia University and a key figure in Koobrik, observed, "Studios have been making screenplays for over 100 years now, which means there is a lot of treasure under their roof that has been forgotten about. Wonderful stories that people poured their hearts into. We want to resurrect these forgotten dreams and ensure that no script is ever forgotten again." This sentiment echoed the deep-seated challenge they aimed to address.

The issue wasn't just about unread scripts; it was about the industry's inability to keep up as scripts piled up, resulting in the opportunities for new and diverse voices to be heard dwindling. The traditional methods of screenplay analysis and selection were proving inadequate in the face of rapidly changing market conditions and evolving storytelling formats.

The challenge was crystal clear; Hollywood's glacial evolution need a revolution. Koobrik needed to leverage technology to cut through this dense thicket of data, unearth hidden narratives, and democratize the process of screenplay selection. The obvious solution was to harness the power of AI and cloud computing to transform the way Hollywood interacts with its most fundamental asset – the screenplay

Why Cloud303?

  • Expertise in AI/ML Solutions Cloud303 possesses in-depth knowledge and expertise in a wide range of machine learning algorithms and artificial intelligence models. Whether it's natural language processing, computer vision, or predictive analytics, Cloud303 is equipped to design, train, and deploy models that deliver actionable insights and drive business value.
  • Ethical and Responsible AI Ethical considerations in AI/ML are crucial, ranging from bias mitigation to data privacy. Cloud303 adheres to ethical guidelines and best practices in AI, ensuring that models are not only efficient but also fair, transparent, and responsible.
  • Scalable Data Processing Managing the massive datasets that feed AI/ML models is a significant challenge. Cloud303 provides scalable data processing solutions, optimizing both storage and computational capabilities. This ensures that your AI/ML models are trained efficiently and can scale seamlessly with your data requirements.
  • Proven Track Record Whether it's navigating complex data migrations, implementing scalable AI/ML models, or setting up robust DevOps pipelines, Cloud303 has consistently demonstrated its ability to deliver, making it a go-to partner for businesses with complex technical needs.

Engagement Overview

Cloud303's engagements follow a streamlined five-phase lifecycle: Requirements, Design, Implementation, Testing, and Maintenance. Initially, a comprehensive assessment is conducted through a Well-Architected Review to identify client needs. This is followed by a scoping call to fine-tune the architectural design, upon which a Statement of Work (SoW) is agreed and signed.

The implementation phase kicks in next, closely adhering to the approved designs. Rigorous testing ensures that all components meet the client's specifications and industry standards. Finally, clients have the option to either manage the deployed solutions themselves or to enroll in Cloud303's Managed Services for ongoing maintenance, an option many choose due to their high satisfaction with the services provided.

The Solution

The Challenge Addressed

At the helm of innovation in screenplay analytics, Koobrik confronted an industry landscape drowning in raw, unprocessed screenplay data. The daunting challenge was not simply to sift through this data but to distill it into secure, scalable, and actionable insights with a layer of efficiency that the industry had not seen before.

Cloud303 architected a solution with AWS's robust services at the core, establishing a digital foundation that began with Amazon S3 as the primary repository. This service became the starting block for Koobrik's screenplay transformation journey, where each script awaited the alchemy that AWS could provide.

The AWS Backbone Enhanced with Bedrock and LLama 2

Cloud303's devised workflow, orchestrated through AWS Step Functions, initiated the transformative screenplay process. The process commenced with Amazon S3 data events, which triggered the subsequent automated steps. Amazon Textract played a pivotal role, meticulously extracting the raw text from various screenplay documents, a process fundamental to the subsequent layers of analysis and synthesis.

The text generation capabilities of Koobrik's platform were significantly amplified by the integration of Amazon Bedrock, paired with LLama 2, which together provided the power to generate not just summaries but intricate character narratives that retained the nuance and complexity of the original scripts. These narratives were crucial for providing context and depth to the analytics provided by Koobrik, delivering content that was rich and engaging for users.

While Amazon Titan 2's embeddings brought a sophisticated level of understanding to the screenplay content, it was not just for refining search capabilities. Instead, the true prowess of Titan 2 lay in its ability to provide nuanced, multi-dimensional representations of screenplay elements, which were instrumental for complex analytical tasks. These tasks could range from understanding character development across a screenplay to identifying thematic patterns within large datasets. Such detailed embeddings allowed Koobrik to unlock new layers of screenplay intelligence, creating a more comprehensive analytical platform.

Data Transformation and Analysis

Following extraction, the data was ushered into an ETL phase, where the raw text was refined and enriched. AWS Comprehend offered advanced content analysis, deconstructing narrative elements to unearth themes, character arcs, and sentiments, further complemented by the contextual depth provided by Titan 2 embeddings.

Building the Data Model

Structured results from Comprehend and enriched embeddings from Titan 2 laid the groundwork for the next phase. Amazon SageMaker stepped in to facilitate the construction, training, and deployment of sophisticated machine learning models capable of handling the complexity and scale of Koobrik's analytical demands.

Enhanced Search and Discovery with OpenSearch

Armed with a rich metadata repository, Koobrik integrated Amazon OpenSearch, which, when paired with the semantic capabilities of Titan 2, offered a powerful script retrieval system. This enabled nuanced ML tasks, such as detailed box office projections and precise actor-script matching, harnessing the collective power of AWS's analytical tools.

Scaling with SageMaker

Amazon SageMaker's endpoints were pivotal in providing scalable, dedicated resources for Koobrik's expanding suite of machine learning tasks. These endpoints allowed for flexible scaling and adaptation, catering to the development of new features and accommodating the growing demands of Koobrik's clientele.

Synchronization and Storage

The Aurora Postgres database ensured seamless data integrity and synchronization across the AWS services, working in tandem with S3's robust storage solutions to preserve the integrity and availability of screenplay data for both immediate and long-term analytical endeavors.

Security and Compliance

Security remained a cornerstone of Koobrik's AWS solution, with AWS Cognito providing robust authentication mechanisms to safeguard system access. Koobrik's deployment also maximized AWS's top-tier security features, ensuring the confidentiality and protection of the unique screenplay datasets.

Engineer Quote

Koobrik's innovative ecosystem, powered by Amazon Bedrock, Textract, Comprehend, and SageMaker, has redefined how screenplays are developed and evaluated. We've not only unlocked dormant stories but also empowered the entire industry to explore new horizons in cinematic storytelling.

Tim Furlong Principal Solutions Architect, Cloud303


The deployment of Koobrik's solution on AWS has resulted in tangible and measurable outcomes that underscore the impact of this technological synergy. Here are the key metrics and success indicators:

Reduction in Processing Time: Screenplay analysis, which previously could take weeks per script, was reduced to a matter of hours. On average, there was an 85% decrease in the time required to process and analyze a screenplay, enabling a faster turnaround for decision-making processes.

Accuracy of Content Analysis: With the integration of AWS Comprehend, the precision of content analysis improved significantly. Metadata extraction accuracy rates soared by 95%, ensuring more reliable insights into themes, sentiments, and narrative structures.

Increase in Script Coverage: Koobrik’s solution enabled a 100% coverage rate in screenplay analysis, ensuring that no script goes unread, compared to the industry standard where an estimated 80% of scripts were not analyzed.

Cost Efficiency: Operational costs were reduced by approximately 60% due to the automation of the screenplay analysis process and the reduced need for manual data entry and analysis.

Scalability Metrics: Koobrik’s AWS-powered solution could seamlessly scale during peak periods, managing a 300% increase in screenplay uploads without compromising on performance or speed.

Security Enhancements: Implementation of Amazon Cognito and AWS security best practices resulted in zero security breaches, maintaining the integrity and confidentiality of sensitive screenplay data.

Innovation Index: Koobrik filed for five new patents related to their screenplay analysis technology within the first six months of AWS implementation, showcasing the innovative edge provided by the AWS suite of services.

These outcomes not only demonstrate the efficacy of Koobrik's AI-powered solution on AWS but also illustrate a paradigm shift in the industry's approach to screenplay analysis. The marriage of AWS's powerful cloud capabilities with Koobrik’s innovative platform has set a new industry standard, transforming the narrative landscape of Hollywood.

On average, there was an 85% decrease in the time required to process and analyze a screenplay, enabling a faster turnaround for decision-making processes.