Graas powers real-time retail intelligence with AWS

Graas is an AI-native decision intelligence platform designed to help commerce brands act in real time across every channel, market, and campaign. From inventory allocation to ad spend optimization, Graas aims to simplify complex retail operations to help brands grow faster and operate smarter.

By running fully on AWS and leveraging AI solutions, Graas has scaled its customer base 3x without increasing infrastructure costs. The platform handles 20x sales surges during major events, while enabling innovations like handwritten order conversion and AI-driven forecasting. With AWS, Graas is delivering instant, data-backed decisions at scale and helping brands stay ahead in a fast-moving digital commerce landscape.

Unifying commerce data into instant decisions

Graas, short for ‘Growth as a Service’, helps digital brands work better across the commerce lifecycle. “Our customers are selling on dozens of storefronts, from their own dotcoms to marketplaces like Amazon and Shopee,” says Prem Bhatia, Co-founder and CEO of Graas. “They advertise on Facebook, Google, and TikTok. They store inventory in multiple warehouses and work with a range of last-mile delivery partners. We bring all of that data together and power decisions in real time; decisions that either grow revenue or protect margin.”

The company works with a diverse mix of clients ranging from global consumer brands to direct-to-consumer startups across fashion, beauty, electronics, and healthcare.

20x scalability without 20x complexity

Digital commerce doesn't follow a steady rhythm. Sale periods in markets like India and Southeast Asia often trigger 20x demand spikes, followed by quiet lulls just days later. That level of volatility requires a highly elastic backend infrastructure.

“Commerce is peculiar. There are certain days where you have to scale up 20x and then shrink back down. AWS lets us do that automatically,” says Bhatia. “We would be struggling if we didn’t have a reliable infrastructure on AWS.”

Graas started out on and continues to build its company with support and infrastructure from AWS. That decision has helped the startup triple its customer base without increasing infrastructure costs, all while maintaining performance on peak demand days.

Moving beyond infrastructure into co-innovation

Over time, Graas’s relationship with AWS evolved from foundational infrastructure to strategic co-development. Today, the AWS team supports Graas not only with technology, but with day-to-day collaboration across architecture and innovation.

“We don’t see AWS as a vendor,” says Bhatia. “The biggest difference is their people. Whether it’s account managers, technical folks, or solution architects, they are genuinely invested in making our business and our customers’ business better.”

That support gives Graas the confidence to move fast and explore new technology frontiers. “The biggest advantage with AWS is the ability to push boundaries without worrying about implementation,” he adds. “This close collaboration with AWS directly enabled us to bring Agent Foundry, our platform for building vertical AI agents, to market.”

Turning scribbles into sales with generative AI

Graas began using with Amazon Bedrock and Amazon Nova to build AI-powered experiences for retail and supply chain. What started as AI use cases have now evolved into agentic AI through Agent Foundry, where agents move from analysis to autonomous action. In one standout use case, Graas transformed handwritten lists into instant digital orders.

“Grocers and pharmacists were uploading lists scrawled by hand. We used Amazon Textract for optical character recognition and fine-tuned Amazon Bedrock models to process those into structured orders in real time,” explains Bhatia. “That’s where generative AI shines in turning unstructured data into immediate action.”

The company is also developing conversational interfaces using Amazon Nova. “Imagine being able to ask, ‘What’s my demand forecast for next quarter?’ and getting that answer back instantly. That’s where we’re heading,” he says.

From raw data to real-time impact

Under the hood, Graas’s stack handles everything from ingestion to inference. ETL pipelines pull in fragmented data, and AI models surface insights that customers can act on immediately.

“Innovations such as Agent Foundry are only possible because of the foundation we’ve built with AWS. As our agent deployments grow, this foundation is absolutely critical,” says Bhatia.

As digital commerce becomes more real-time, more data-rich, and more competitive, Graas is helping brands meet that future head-on. With the support of AWS, the company is setting a new standard for intelligent decision-making at scale.

Startups like Graas are able to grow faster while still managing tight budgets and limited resources with the support of AWS Activate. In addition to technical guidance and tailored tools, the program offers startups up to US $100,000 in AWS Activate Credits which can be used to offset the cost of infrastructure, data services, and AI and ML models. Since its inception in 2013, the program has provided more than $8 billion in credits to founders across the world. Join AWS Activate today and discover how to unlock your credits and access an extensive set of resources.