Case study
Five Scanning Tools Unified Into One Source for a Semiconductor Maker
How SquareShift built a Vertex AI chatbot on one BigQuery dataset that unifies five scanning tools, giving security analysts direct answers instead of manual cross-checking.
Book a session →
Client
A company that designs, makes, and supplies semiconductor and networking hardware for data centers and industrial customers.
Industry
Technology
Services
Stack
Vertex AI · Gemini Pro · PaLM 2 · BigQuery · RAG (Retrieval-Augmented Generation) · Armorcode · BlackDuck · Lacework · JFrog · Coverity · Qualys
How SquareShift delivered it.
The challenge
Security analysts had to check vulnerability findings by hand across five separate scanning tools and outside websites, with no single place to get a contextual answer instead of cross-checking everything themselves.
What we delivered: a Vertex AI chatbot on one BigQuery dataset
SquareShift built a vulnerability analysis chatbot on Vertex AI that answers analyst questions directly, backed by one BigQuery dataset that combines external advisories with Armorcode data instead of keeping them apart in separate systems.
The outcome
Analysts now get contextual answers from the chatbot instead of checking each tool by hand, and remediation suggestions and status updates are generated in bulk instead of one at a time.
A chatbot that hallucinates a fix is worse than no chatbot at all. Getting the retrieval right, grounded in the same data analysts already trust, is what makes the answers usable instead of just plausible.
AI & Machine Learning Practice Lead, SquareShift
Related reading: HR chatbot on Vertex AI and UX friction analysis with BigQuery and Gemini.
