Ringg AI voice agents now handle 20 million call attempts a month in India, and the startup has just secured a $10 million extension to its Series A from Peak XV Partners to push into territory that feels considerably harder than a loan-collection bot.
The new capital brings the total Series A to $15.5 million. The first close of $5.5 million, announced on the Ringg AI company blog, was led by Arkam Ventures, with Groww Founder Fund, Kunal Shah, Whitecap Ventures, and Capital2B also participating. Peak XV Partners, the fully independent firm that spun out of Sequoia India and Southeast Asia and launched with $2.5 billion of uninvested capital, is writing the cheque for round two.
From Cheap Calls to Complex Workflows
Ringg did not start here. Co-founders Siddharth Tripathi, Utkarsh Shukla, and Kali Charan Vemuru, listed on the Ringg AI careers page, first built a text-to-speech business called DesiVocal. Training proprietary speech models proved too expensive at that layer, so the team moved up the stack into voice AI agent software for enterprises. DesiVocal officially closed on 30 July 2026, its site redirecting users to a Ringg contact address.
Indian fintech Cred was the first enterprise customer. Flipkart, Practo, Groww, and PolicyBazaar followed. Early use cases were high volume and low margin: outbound calling, lead qualification, loan collection. Tripathi was candid about why that model ran out of road.
‘At the start, we were doing high-volume, low-complexity use cases like outbound calling, lead qualification, loan collection, and more. We quickly realized these are not sticky use cases, and so it’s always going to be a price game,’ he told TechCrunch.
Ringg AI voice agents now run across 1,200 clinics on the healthcare platform Practo, scheduling appointments and following up with patients after visits. Abandoned-cart recovery for e-commerce and onboarding and KYC checks for fintech apps represent the same push toward workflows that are harder to replicate and harder to reprice. For clients including Shell, the platform also automates browser-based support requests.
Voice calls still account for over 70% of Ringg’s revenue, but the company has extended into chat and WhatsApp. ‘We are trying to position ourselves as a platform for agents that bring outcomes or get things done rather than voice agents for enterprises,’ Tripathi said.
The Technical Moat Ringg Is Building
The engineering ambition is visible in the product. In June 2026 Ringg introduced a proprietary speech-to-text model called Parrot, built specifically for real-time voice AI with support for Hindi-heavy calls, low latency, and production-grade accuracy. The longer-term goal is to own the full voice stack, from models through to infrastructure and deployment. For now, costs make that impractical, so the platform operates as an orchestration layer, routing tasks to whichever model best fits the use case.
Pricing reflects that pragmatic structure. Ringg AI’s industries page lists a flat fee starting at approximately $0.08 per minute, bundling large language model, voice, and telephony costs into a single line item for enterprise buyers. The company is also certified as SOC 2 Type II, ISO 27001, HIPAA, and GDPR-compliant, a credential set that matters when the workflows involve patient data or financial identity checks.
Rishen Kapoor, a principal at Peak XV, credited Ringg’s research-lab origins for its ability to handle the messier end of enterprise automation. ‘Because of the technical capabilities, they can actually do these hard-won enterprise workflows end to end. They can complete these higher-value tasks like merchant onboarding, like L1 and L2 support, with quality and with consistency,’ Kapoor told TechCrunch.
A Crowded Field and a Deliberate Geographic Bet
The competitive landscape is layered. Model makers such as Deepgram, ElevenLabs, Cartesia, Sarvam, and Smallest.ai compete at the foundation level. Orchestration-focused startups Bolna and Blue Machines target the same middleware layer Ringg occupies. Sector specialists like Gnani and Arrowhead concentrate heavily on finance. In a market this structured, the defensible position belongs to whoever owns the end-to-end customer outcome, not just one component of the call.
Ringg’s geographic strategy is equally deliberate. Most customers are in India, with a handful in the Middle East and the US. Rather than selling directly to American enterprises, the company wants to partner with Global Capability Centers, the offshore hubs that multinationals use for back-office and support work, to offer automation capacity alongside human agents.
The company currently employs 40 people, with more than 15 joining in the past three months. Open roles include forward-deployed engineers who combine technical and product management skills, and researchers focused on reducing model-running costs. That last hire category is the one to watch: if Ringg can bring down inference costs, the full-stack ambition stops being a long-term aspiration and starts being a near-term build.
