OnFinance raises 4.2 million in pre-series A for BFSI-focused Artificial Intelligence

Bengaluru-based OnFinance raised 4.2 million in a pre-series A round led by Peak XV’s Surge. The startup plans to scale in India, expand to the US and MENA, and deepen BFSI-specific large language model research.

OnFinance has raised 4.2 million (about INR 37 crore) in a pre-series A round led by Peak XV’s Surge. The round saw participation from Groww Founders’ Fund, MarsShot VC (Razorpay Founders’ Fund), Climber Capital, and Shyamal Hitesh Anadkat, head of applied Artificial Intelligence at OpenAI, alongside existing backers Indian Angel Network and Silverneedle Ventures. The Bengaluru-based startup said it will use the capital to scale contracts with Indian clients, foray into the US and MENA markets, and double down on BFSI-specific large language model research.

Founded in 2023 by Anuj Srivastava and Priyesh Srivastava, OnFinance began by offering Artificial Intelligence copilots for equity research analysts before pivoting to compliance, risk, and audit. The company argues that compliance datasets such as SEBI circulars and RBI guidelines are more structured and closed ended, making them better suited for enterprise-grade Artificial Intelligence than open-ended research tasks. This shift led to the launch of ComplianceOS, an operating system for compliance teams powered by NeoGPT, the company’s proprietary BFSI-focused large language model built on top of open source Llama 3.3 and trained on decades of SEBI, RBI, IRDAI, and AMFI data.

OnFinance has also developed InvestigativeOS, which hosts more than 70 Artificial Intelligence agents for risk and audit workflows. These agents automate tasks such as KYC validation, SEBI’s cyber resilience compliance, vendor risk assessments, accessibility compliance, and market abuse surveillance covering insider trading, front running, and manipulative trades. The startup counts BSE, Kotak Mutual Fund, Nippon Mutual Fund, HDFC Securities, and Aditya Birla Capital Digital among its more than 15 clients. It said it was EBITDA positive in March 2025 and nearly breakeven for FY25, with an approximately INR 74 lakh loss for the year.

The company is piloting its platform with banks in the US and expects deal sizes to be two to three times larger than in India, with strong upsell potential across banking, brokerage, and insurance units. OnFinance deploys its stack on premises by helping institutions procure GPUs for bare metal servers, or via private clouds on AWS, Azure, GCP, or Oracle. It has also built a Model Context Protocol server that connects its agents with clients’ systems of record such as HRMS and internal governance tools, which the company says enables production-grade deployments. The raise comes amid accelerating adoption of Artificial Intelligence in fintech, with other agentic and no-code players like Pascal AI and iTuring.ai also attracting investor interest in recent months.

50

Impact Score

Nvidia to sell fully integrated Artificial Intelligence servers

A report picked up on Tom’s Hardware and discussed on Hacker News says Nvidia is preparing to sell fully built rack and tray assemblies that include Vera CPUs, Rubin GPUs and integrated cooling, moving beyond supplying only GPUs and components for Artificial Intelligence workloads.

Navigating new age verification laws for game developers

Governments in the UK, European Union, the United States of America and elsewhere are imposing stricter age verification rules that affect game content, social features and personalization systems. Developers must adopt proportionate age-assurance measures such as ID checks, credit card verification or Artificial Intelligence age estimation to avoid fines, bans and reputational harm.

Large language models require a new form of oversight: capability-based monitoring

The paper proposes capability-based monitoring for large language models in healthcare, organizing oversight around shared capabilities such as summarization, reasoning, translation, and safety guardrails. The authors argue this approach is more scalable than task-based monitoring inherited from traditional machine learning and can reveal systemic weaknesses and emergent behaviors across tasks.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.