Vaani AI Research, a foundation speech model company based in Bengaluru and Singapore, today announced the launch of its self-serve developer platform for building, deploying and managing voice AI agents. The platform enables developers, startups and product teams to create agents through APIs or a configurable voice harness, without having to assemble and manage the underlying infrastructure themselves.
Teams can describe the agent they want to build, preview its voice, configure its actions and deploy it across phone calls, websites, apps and WhatsApp. Developers can control how agents respond, maintain conversational context, use tools and transfer conversations to human teams.
At the core of the platform is Vaani’s voice harness, an infrastructure layer that connects its speech models with the systems needed to run agents in production. It brings together inference optimisation, memory and context management, conversational orchestration, telephony, tool integrations and human-in-the-loop workflows. The launch addresses a recurring challenge in voice AI: moving from a successful demonstration to reliable performance in real conversations. Agents must handle callers switching between languages mid-sentence, retain relevant context and respond to unexpected inputs. Teams also need flexibility to refine these interactions without relying solely on extensive prompt engineering. Vaani’s platform brings these capabilities into a single stack, making agents easier to build while giving developers precise control over the experience they deliver. “Seventy-five percent of the world does not primarily speak English, and most voice AI still treats that as an edge case. We built this harness inside live enterprise deployments, where every call shows you what breaks. Today we are opening it to anyone who wants to build a voice agent in the language their customers actually speak, said Tushar Shinde, Co-founder and CEO, Vaani AI Research.
Vaani’s self-serve platform emerged from its work deploying voice AI across enterprise workflows, where poor conversational experiences can lead to lost opportunities and frustrated customers. Drawing on those deployments, the company developed its own speech models, inference infrastructure and orchestration layer to address how agents listen, respond, act and retain context. Opening this infrastructure to a broader developer community allows teams to focus on their applications and customer experiences while Vaani manages the underlying voice stack.The platform supports both independent development and more complex enterprise implementations. Developers and product teams can configure and deploy agents through the self-serve harness, while enterprises with specialised requirements can work with Vaani’s forward-deployed engineering team on integrations and workflow optimisation. Vaani’s infrastructure is already being used by businesses in India and international markets. With the self-serve launch, the company aims to make production-grade voice AI more accessible to teams building multilingual customer interactions.