Running on Google Cloud, Replenit sits on top of a brand's existing technology stack, automatically deciding which shopper to contact, when to reach them, and what context to use. Instead of forcing marketing teams to build campaigns manually, the engine creates ready-to-use instructions and sends them directly to a brand’s communication channels.

Replenit relies on BigQuery to gather, normalize, and enrich complex customer data and product details into unified records. For cost-effective, high-volume processing, Replenit feeds this data into open-source Gemma models that it hosts and fine-tunes on Tensor Processing Units (TPUs). This allows the system to generate rich, customized data insights in minutes.

The Replenit team used Gemini Enterprise Agent Platform to build a multi-agent workflow where distinct, complementary AI agents take on specialized roles. Individual agents manage specific pipeline stages—such as enriching limited, fragmented retail data into structured memory for the customer, the product, and the brand. It then applies the industrial skill matched to the KPI you're optimizing for, and owns the workflow end-to-end: replenishment, cross-sell, engagement, churn, and winback, for every retailer on the platform. The platform uses Gemini Pro for deep contextual reasoning and Gemini Flash for high-speed, volume-heavy execution.

“We built an engine that runs hundreds of specialized agents simultaneously to read and enrich retail data,” explains Egemen Akdan, co-founder and CTO at Replenit. “Using BigQuery and Gemini Enterprise Agent Platform, we can cleanly orchestrate these complex agentic workflows, turning raw behavioral signals into precise, actionable intent.”

For the consumer, this means they receive the right information at exactly the right moment, such as a helpful suggestion tailored to their specific situation. For instance, a shopper might see a compatible skincare item suggested precisely when their routine requires a restock. Meanwhile, supported by built-in compliance from Google Cloud, which carries certifications such as SOC 2, Replenit easily implements the data separation, tenant isolation, and auditable decisions that enterprise retailers demand.

By optimizing data operations through Google Cloud, Replenit has unlocked significant operational improvements. With Gemini Enterprise Agent Platform Pipelines, the team has reduced pipeline execution times by more than 10x. Combining self-hosted open models with services from Gemini also cut overall pipeline costs by 10x, from over $200 to just $20 per run. This 90% cost reduction allows resources to be reinvested into product innovation. New production-ready pipelines can now be developed and shipped in as little as four hours instead of three days, meaning the product team can test new models and run experiments continuously, rather than waiting days for results.

These operational improvements, combined with the ability to clear stringent enterprise security checks, have allowed Replenit to scale rapidly. The company achieved a 360% increase in business growth over a single quarter because it could confidently onboard enterprise retailers at scale. Furthermore, because the platform can enrich data flexibly without industry limitations, Replenit expanded its total addressable market immediately across all retail types, skipping the slow path of moving vertical by vertical.