AI-powered product data management and syndication for e-commerce
Trustana operates a product information management platform aimed at retailers and distributors who struggle with fragmented, incomplete product catalogs. The tech stack reveals a machine-learning-first architecture: PyTorch, TensorFlow, and multiple LLM providers (OpenAI, Anthropic, Hugging Face, Qwen) paired with vector databases (Pinecone, Weaviate, FAISS) and RAG pipelines — indicating heavy investment in AI-driven enrichment and classification. Active projects centered on LLM workflows, inference pipeline scaling, and catalog automation align with their stated pain points around data quality and managing product data at scale.
Trustana is a SaaS platform for managing and enriching product data across e-commerce channels. Founded in 2020 and seeded by Temasek, the company serves mid-market and enterprise retailers, distributors, and brand owners who need to unify, enrich, and syndicate product information across multiple sales channels. The product handles the full product data lifecycle: aggregating first-party data, filling information gaps with AI, generating localized marketing content, and enabling cross-channel syndication. The team of 11–50 is Singapore-based with hiring activity across Malaysia, Vietnam, Japan, Canada, the United States, and Australia.
Trustana builds on Python and TypeScript with PyTorch and TensorFlow for ML, integrates multiple LLM providers (OpenAI, Anthropic, Hugging Face, Qwen), and uses vector databases including Pinecone, Weaviate, and FAISS. Testing relies on Playwright, Cypress, and Selenium with CI/CD automation.
Active projects include LLM-based product classification and enrichment, scalable AI inference pipelines and APIs, automated test framework development, CI/CD automation, and go-to-market strategy refinement. Internal focus areas are test reliability, inference pipeline optimization, and product data quality at scale.
Other companies in the same industry, closest in size