Use Case Intelligence RSS Search use cases Search
AWS Healthcare Global

Vertex Pharmaceuticals

Application Refactoring · Enterprise Knowledge Search · IT Operations Automation · Real-time Analytics · Security and Compliance

After designing an experiment, they manually trained the base ML model about what they hoped to identify from the images, labeling each area of interest by hand. In the legacy system architecture, all scientists shared a single pool of compute resources.

Read the original story Explore all use cases
11x accelerated image segmentation
90% cost reduction
Reducing Processing Time by 11x and Cost by 90% Using AWS Step Functions at Vertex Pharmaceuticals
Using AWS Step Functions Distributed Map for large-scale parallel data processing, Vertex has accelerated experiment image analysis by 11 times while reducing costs by 90 percent, helping scientists iterate on their research more quickly in their search for transformative therapeutics.

Reported impact

Business outcome

Vertex has identified other areas where it can apply its learnings from AWS Step Functions Distributed Map. For example, during the analysis of a large library of scientific compounds, one unit of work spawns parallel units that need to be aggregated. “Using AWS Step Functions, we can tackle our scientists’ issues,” says Iturralde. “We’re here to accelerate research and help bring safe, effective medicines to patients quickly.” Using AWS Step Functions, we greatly improved the robustness and reliability of our image segmentation solution. Karthik Ghantasala Director of Cloud Architecture and Strategy, Vertex Pharmaceuticals

  • Cost Reduction
  • Productivity Improvement
  • Risk Reduction
  • Scalability

Solution

Vertex Pharmaceuticals (Vertex) is a global biotechnology company that invests in scientific innovation to create transformative medicines for people with serious diseases. To aid scientific research, scientists use image segmentation, a technique that parses pictures into discrete groups of pixels and uses machine learning (ML) to identify areas of interest. Using AWS Step Functions, Vertex accelerated performance by 11 times because workflows now take minutes instead of hours. “Initially, some of our scientists questioned if the system was working, because it was going so fast compared with the old one,” says Iturralde. Using AWS Step Functions Distributed Map for large-scale parallel data processing, Vertex has accelerated experiment image analysis by 11 times while reducing costs by 90 percent, helping scientists iterate on their research more quickly in their search for transformative therapeutics. Using the scalability of AWS Step

Data flow

workflow control is handled by AWS Step Functions, AWS Step Functions Distributed Map; processing runs on AWS Fargate, AWS Lambda; results and state are stored in Amazon S3; and outputs are exposed through Amazon API Gateway for the image segmentation workload.

Solution architecture

7 components · 5 layers
  1. Compute
    • AWS Fargate Runs core compute or model execution for the image segmentation workload.
    • AWS Lambda Runs core compute or model execution for the image segmentation workload.
  2. Storage
    • Amazon S3 Stores application, analytics, or workflow state for the image segmentation solution.
  3. Serving
    • Amazon API Gateway Provides the delivery or user-facing interface for the image segmentation workload.
  4. Orchestration
    • AWS Step Functions Coordinates the image segmentation workflow and automation steps across services.
    • AWS Step Functions Distributed Map Coordinates the image segmentation workflow and automation steps across services.
  5. Governance
    • Amazon CloudWatch Adds monitoring, governance, or operational controls around the image segmentation deployment.

Architecture clues

  • Scientists interact with a web application running on AWS Fargate , a serverless, pay-as-you-go compute engine that lets engineers focus on building applications without managing servers.
  • The AWS Step Functions workflow uses the input from Amazon API Gateway to determine which images to process and which parameters to pass back to the ML model.
  • The automated process begins with a call to Amazon API Gateway, a fully managed service that makes it simple for developers to create, publish, maintain, monitor, and secure APIs at nearly any scale.
  • Using AWS Step Functions Distributed Map, the system processes all the images in parallel.
  • Using AWS serverless technologies, the subsequent batch of images now process automatically.

Evidence from the source

After designing an experiment, they manually trained the base ML model about what they hoped to identify from the images, labeling each area of interest by hand.
In the legacy system architecture, all scientists shared a single pool of compute resources.
Scientists interact with a web application running on AWS Fargate , a serverless, pay-as-you-go compute engine that lets engineers focus on building applications without managing servers.
The automated process begins with a call to Amazon API Gateway, a fully managed service that makes it simple for developers to create, publish, maintain, monitor, and secure APIs at nearly any scale.
Using AWS serverless technologies, the subsequent batch of images now process automatically.
Using the web application, scientists fine-tune the parameters of the ML model that segments the image and test the model on individual images.
Transparency Evidence & methodology 95/100 reference value · 95% confidence · 2 sources

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value95/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth20/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    6 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence95%
  • Source authority25/25
    The deployment is described by the adopting organization or a government source.
  • Claim traceability23/25
    11 of 12 key fields link to a named source.
  • Direct evidence20/20
    6 direct source quotes retained.
  • Date provenance8/10
    The publication date comes from publisher-controlled metadata.
  • Source transparency9/10
    2 source references document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. AWS

    Vertex Pharmaceuticals customer story

    First-party
    https://aws.amazon.com/solutions/case-studies/vertex-pharmaceuticals-distributed-map/
    Source type
    Vendor-published customer story
    Published
    Oct 29, 2024
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date
  2. U.S. Securities and Exchange Commission

    Vertex Pharmaceuticals Form 10-K for the fiscal year ended December 31, 2025

    Government
    https://www.sec.gov/Archives/edgar/data/875320/000087532026000056/vrtx-20251231.htm

    The filing states: "we had approximately 6,400 employees."

    Source type
    Government or regulatory registry
    Supports
    • Company size

Date evidence

Not separately sourced Industry, Region. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
AWS Sep 7, 2023 Bio-Rad Laboratories Healthcare Global Additionally, the team needed Internet of Things (IoT) capabilities to facilitate the automatic and programmatic upload of data—including high-volume instrument data and low-volume system telemetry data—to a centralized… A single ML model might analyze up to 50 GB of data, making it impractical to run on a data scientist’s laptop. Application RefactoringEnterprise Knowledge Search AWS Glue · AWS IoT Greengrass · Amazon S3 AWS Apr 17, 2024 Chunghwa Telecom Telecommunications Taiwan Furthermore, Chunghwa Telecom needed a platform with robust security controls. To improve artificial intelligence (AI) data security and governance, the company migrated its generative AI projects to AWS. The first generative AI solution, the SDLC assistant, has saved developers an estimated 13 percent of their… Document ProcessingEnterprise Knowledge Search Amazon Bedrock · Amazon Polly AWS Mar 5, 2024 Mercedes-Benz Consulting Commercial and Professional Services Global With millions of internal documents to search through, getting relevant answers quickly was a standing challenge for employees. Additionally, if an employee left the company or a colleague with specific knowledge was… Application RefactoringDocument Processing AWS Glue · AWS Lambda · Amazon OpenSearch Service