As more customers began to launch their applications with ATMA, Adya started to face development roadblocks. The DevOps team was building applications with standard cloud practices, which meant it could take two to six months to launch an app.

“We expressed this problem to the Google Cloud team, and they provided us with a solution to automate the entire scripting process,” Mazumder says. Using Terraform on Google Cloud, ATMA automates the script and launches an app in just one hour instead of months. It also standardizes the product lifecycle and codebase for better reliability and security. 

As a fully self-service platform, Adya relies on Google Kubernetes Engine (GKE) to launch a scalable architecture. This allows customers to launch their apps and scale at any time, while keeping costs low. But as Adya's user base grows, so does the amount of data flowing through the platform. This high volume of data can quickly overload the system, but with GKE, Adya can pull together tools like Kafka, Redis, and Elastic search to manage queuing, streaming, and caching at all times.

Apart from ATMA, Adya also developed a custom large language model (LLM), ADI (Adya Deep Intelligence), and an agent orchestration platform called Vanij. While ADI consists of multiple proprietary models, the Adya team collaborates with Gemini for Google Cloud to create various use cases around it. “We recently created a plugin that will help people perform voice searches on our applications, and that voice search leverages Gemini,” Mazumder explains.

By layering ADI and Gemini, Adya is working on enterprise search, training the LLM so users can simply ask a question and get a generated response.

From the day Adya launched, Mazumder knew the company would use Google Workspace for its day-to-day operations because of its ease of use, security, and accessibility. And with the rest of Adya’s operations running on Google Cloud, adopting workspace allowed for seamless interoperability and collaboration at scale.

Hard work and collaboration have paid off. Having worked closely with Google Cloud to build its products since inception, Adya now has six solutions on the Google Cloud Marketplace. These solutions include an inventory seller app, marketplace seller app, finance and reconciliation management app, logistics app, issues and grievances management app, and a buyer app. 

One year since launch, Adya has big plans for the near future. “My biggest focus now is to combine Vanij and ATMA, and launch entirely on Google Cloud,” Mazumder says. “I believe we will witness even bigger growth and scalability as more users come on board. By continuing to work closely with the team at Google Cloud, we will go far.”