Supply Chain Optimization
Sourcy needed to turn rough buyer requests into manufacturable product specifications and supplier quotes quickly, but traditional product-development handoffs depended on slow manual coordination that was difficult to scale across languages and suppliers.
Reported impact
Sourcy now turns buyer requests into factory-ready specifications in seconds, reduced blended AI cost to roughly $1 per million tokens, and expanded its production estate from two to 23 Cloud Run services while processing around four billion tokens per month.
Sourcy built a multi-agent prompt-to-product platform on Google Cloud so consumer brands can turn rough product ideas into supplier-ready specifications in seconds instead of hours. The system uses Gemini Enterprise Agent Platform to orchestrate specialized agents for customer activation, specification building, supplier discovery, supplier engagement, and quote preparation. Through the Gemini API, Sourcy routes high-volume work to lighter models and complex reasoning to Gemini Pro while keeping AI costs low at large token volumes. Cloud Run hosts the company's expanding set of containerized services, while Cloud SQL, Firestore, Pub/Sub, BigQuery, Document AI, Speech-to-Text, App Engine, Google AI Studio, and Compute Engine support the surrounding data and processing stack. The result is a multilingual sourcing platform that compresses idea-to-factory workflows and builds a reusable sourcing-intelligence layer from supplier and quotation data.
Buyer descriptions, documents, and spoken inputs enter Sourcy through multimodal ingestion services, flow through Gemini-orchestrated agents for specification building, supplier discovery, and quote preparation, run on Cloud Run-hosted services, and persist supplier and quotation outputs in operational stores and BigQuery for future sourcing reuse.
Deploying these services on Cloud Run also helps the team scale while continuing to manage a heavy transactional load.
Sourcy Brain orchestrates specialized agents for customer activation, specification building, supplier discovery and engagement, and quote preparation.
Sourcy built a multi-agent sourcing system on Gemini Enterprise Agent Platform.
Sourcy was able to grow from two services to 23 with the company’s existing team size.
This model tiering has supported billions of processed tokens at a blended AI cost of roughly $1 per million tokens.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Google Cloud Sample Modified · Sep 17, 2026 Mentana Founded in Australia, Mentana is an AI SaaS platform designed to modernize global supply chain operations. Operating as an AI orchestration layer, the platform unifies real-time reasoning across legacy enterprise… Achieves 99.99% uptime and sub-5-minute scaling using Compute Engine Artifact Registry · Compute Engine · Gemini Google Cloud Sample Archived · Apr 13, 2026 Tictag.io A private endpoint within Virtual Private Cloud (VPC) that connects the company's cloud environment securely and with low latency to its primary MongoDB Atlas database, and To power intensive AI processing for its… "We selected the NVIDIA L4 Cloud GPU because we can fit two into one node and it sits at the sweet spot of cost… Cloud CDN · Cloud GPUs · Cloud Storage Google Cloud Sample Modified · Oct 9, 2026 Luxia Agro Luxia Agro is an Argentine company specializing in the import of phytosanitary products for the country's agricultural sector. With a core team of just 10 people, the company has built a reputation for quality and… 50% reduction in labor hours Cloud Run · Cloud SQL · Cloud Storage 