Agentic AI startup Karta AI has raised US$3 million (approximately ₹280 million) in a Seed funding round led by PeerCapital and Cedar Hill Capital. The round also drew participation from Wyser VC, CDM Capital, and Seedstars. Founded in 2024 by former Razorpay executives Vijaypal Singh and Abhishek Kushwaha, the company plans to use the capital to expand its go-to-market strategy, distribution, and product development over the next two years.
Funding Overview
The Seed round validates Karta AI's approach to automating complex inbound customer service interactions for financial institutions. The startup will prioritize go-to-market execution and distribution over the next two years, with a portion of the funds allocated to emerging AI capabilities such as self-learning. Product development will continue to focus on voice-first experiences, which the company sees as a significant opportunity in the Indian market.
Founding and Market Focus
Vijaypal Singh and Abhishek Kushwaha started Karta AI after leaving Razorpay, where they worked on internal AI automation. The startup initially focused on chat-based automation before expanding into voice in August 2025. It now targets inbound interactions such as failed payments, blocked cards, loan-related queries, and insurance claims, where customers often raise multiple issues in a single call.
Targeting Complex Customer Queries
Karta AI focuses primarily on voice-based interactions and targets level 2 and level 3 queries that require coordination across multiple enterprise systems. These complex queries differ from level 1 FAQs, which can be resolved through a single API call. According to CEO Vijaypal Singh, level 2 and level 3 queries account for roughly 80% of customer service costs even though they represent a smaller share of total volumes.
Customer Base and Operational Metrics
The startup claims to handle around 2 million customer conversations per month across more than 25 languages. It has 10 to 12 paying customers, including PhonePe, Mindlabs, Bajaj, and Aavas, and its agents resolve about 70% of borrower queries without human intervention. Karta AI reports that its platform reduces cost per query by approximately 50% for lending-related requests such as loan disbursal updates, statements of account, and foreclosure statements.
Technology and Workflow Automation
Karta AI's proprietary context engine helps its agents manage multi-intent conversations by identifying customer intent and retrieving relevant enterprise data before generating responses. The engine draws on standard operating procedures, knowledge bases, previous conversations, and API data to handle follow-up requests within the same call. The platform integrates with enterprise systems to execute customer-facing and back-office workflows, using models from OpenAI, Google's Gemini family, and open-source options such as Kimi and Gemma.
Compliance and Future Expansion
Karta AI does not automate every interaction, and queries involving fraud or high-risk matters are routed to human agents. Customers can also request a human representative, while the platform applies compliance guardrails and testing before deploying agents. The company primarily serves Indian businesses in banking, financial services, and insurance, and it plans to develop reusable industry-specific capabilities before expanding internationally.
Karta AI's Seed funding arrives as enterprises seek to reduce customer service costs while handling increasingly complex inbound requests. The startup's focus on level 2 and level 3 voice queries positions it in a segment that carries the majority of support expenses for financial institutions. With new capital, Karta AI aims to strengthen its technology, expand its customer base, and eventually bring its automation platform to international markets.
Source: Inc24