Named clients.
Real work.
Every story names the client, the problem, what we built and how it is governed. Results appear where the client has approved them.
- Client stories
- 23
- Industries
- 7
- Practices
- 4
Claude ships Ava's documentation with the code
- 01Client and problem
Help articles were assembled by hand and drifted away from the product with every release.
- 02What CWX delivered
A Claude workflow inside the release pipeline that writes and checks each article as code ships.
- 03What changed
Documentation now ships with the code, and a second person approves every update before customers see it.
Claude keeps TIMIA's loan book up to date
- 01Client and problem
Loans were tracked by hand, with little record of why a number changed.
- 02What CWX delivered
Claude reads each credit agreement and drafts the loan-book change, citing its source clause.
- 03What changed
Every change is traced to its clause, and a person approves each write.
Moving Rad AI's radiology platform from AWS to Google Cloud
- 01Client and problem
Rapid growth needed a secure, HIPAA-compliant platform for AI-powered radiology.
- 02What CWX delivered
Migration from AWS to Google Cloud, with applications refactored for HIPAA compliance and GPU capacity for model training.
- 03What changed
Rad AI’s platform now runs on Google Cloud, refactored for HIPAA compliance, with the GPU capacity its model training needs.
AI-powered procurement for acceleration
- 01Client and problem
Buyers could not search structured and unstructured supplier data in plain language.
- 02What CWX delivered
Generative AI search on AlloyDB Vector Store and Vertex AI, with contextual filtering.
- 03What changed
Natural-language supplier search that surfaces relevant suppliers faster.
Accelerating drug discovery with revolutionary AI
- 01Client and problem
Drug discovery research slowed by manual data extraction and analysis.
- 02What CWX delivered
Vertex AI and Document AI models, plus a Gemini assistant scientists can question.
- 03What changed
Scientists spend far less time harmonizing data and more on discovery.
Enhanced asset creation propels mobile gaming powerhouse
- 01Client and problem
In-game banners, characters and backgrounds took too long and cost too much to produce.
- 02What CWX delivered
A tuned Stable Diffusion model on Google Cloud, with curated data and a secure interface.
- 03What changed
Faster, lower-cost asset generation that fits Kabam’s existing development pipeline.
Democratizing the mortgage industry with AI solutions
- 01Client and problem
Manual document review slowed every loan through underwriting.
- 02What CWX delivered
A generative AI underwriting copilot on Vertex AI and Dialogflow, using directed retrieval.
- 03What changed
One copilot reviews documents across every loan product, so loan officers decide faster.
Meet BenBot, raising the bar for online capital raising
- 01Client and problem
Customer conversations were spread across HubSpot, Slack and email, with inconsistent answers.
- 02What CWX delivered
BenBot, a conversational hub on Vertex AI trained on a generative playbook, across every channel.
- 03What changed
Faster, consistent customer communication, with real-time monitoring as the platform scales.
Groundbreaking multi-turn Vertex AI conversation agent
- 01Client and problem
Field-service teams needed real-time, AI-driven answers on technicians and operations.
- 02What CWX delivered
A multi-tenant Vertex AI conversation agent built with Dialogflow CX, LangChain and ReAct.
- 03What changed
Personalized support and analysis across structured and unstructured field data.
Outdoor e-tailer wins big with custom personalization
- 01Client and problem
Every shopper saw the same products, whatever their interests.
- 02What CWX delivered
Personalized recommendations built on Google Retail API and BigQuery, served through Vertex AI.
- 03What changed
Product suggestions tailored to each customer’s behaviour and preferences.
Advanced AI making prayer a priority globally
- 01Client and problem
Users needed accurate answers drawn from a vast library of recordings and transcripts.
- 02What CWX delivered
A retrieval-based generative AI assistant on Gemini, grounded in Pray.com’s own content.
- 03What changed
Deployed to the client's development environment, with safeguards and an evaluation system.
Optimizing the in-store shopping experience with AI-powered knowledge management
- 01Client and problem
In-store stylists needed to navigate 7+ systems to manage purchases, inventory, returns, security and more.
- 02What CWX delivered
STITCH, an AI knowledge assistant that answers across knowledge bases and SOPs in seconds, on a RAG workflow.
- 03What changed
An early pilot boosted stylist productivity and reduced time to sale, and Levi’s moved STITCH toward a wider store rollout.
Engineering autonomous marketing asset selection and generation
- 01Client and problem
Selecting and optimizing creative assets for account-based marketing journeys was manual, slow to market and subjective.
- 02What CWX delivered
A dual-agent system that learns from historical CTR data and image features to suggest assets autonomously.
- 03What changed
In five weeks, the creative workflow went from a cost center to a high-velocity performance engine, increasing CTR by 17% and reducing consulting time by 20%.
Let's talk about what's next.
Bring one question or one use case. We will bring the people who can answer it.